index
int64
repo_id
string
file_path
string
content
string
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/RapidsHelpV3.java
package water.api.schemas3; import water.Iced; import water.api.API; /** * Help for the rapids language */ public class RapidsHelpV3 extends SchemaV3<Iced, RapidsHelpV3> { @API(help="Description of the rapids language.", direction=API.Direction.OUTPUT) public RapidsExpressionV3[] expressions; public static class RapidsExpressionV3 extends SchemaV3<Iced, RapidsExpressionV3> { @API(help="(Class) name of the language construct") public String name; @API(help="Code fragment pattern.") public String pattern; @API(help="Description of the functionality provided by this language construct.") public String description; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/RapidsMapFrameV3.java
package water.api.schemas3; import water.Iced; import water.api.API; import water.fvec.Frame; import java.util.Map; public class RapidsMapFrameV3 extends RapidsSchemaV3<Iced,RapidsMapFrameV3> { @API(help="Frames", direction=API.Direction.OUTPUT) public RapidsFrameV3[] frames; @API(help="Keys of the map", direction=API.Direction.OUTPUT) public RapidsStringsV3 map_keys; public RapidsMapFrameV3() {} public RapidsMapFrameV3(Map<String, Frame> fr) { map_keys = new RapidsStringsV3(fr.keySet().toArray(new String[]{})); int i = 0; Frame[] framesFromMap = fr.values().toArray(new Frame[]{}); frames = new RapidsFrameV3[framesFromMap.length]; for (Frame frame : framesFromMap) { this.frames[i++] = new RapidsFrameV3(frame); } } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/RapidsNumberV3.java
package water.api.schemas3; import water.Iced; import water.api.API; /** */ public class RapidsNumberV3 extends RapidsSchemaV3<Iced,RapidsNumberV3> { @API(help="Number result", direction=API.Direction.OUTPUT) public double scalar; public RapidsNumberV3() {} public RapidsNumberV3(double d) { scalar = d; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/RapidsNumbersV3.java
package water.api.schemas3; import water.Iced; import water.api.API; /** */ public class RapidsNumbersV3 extends RapidsSchemaV3<Iced,RapidsNumbersV3> { @API(help="Number array result", direction=API.Direction.OUTPUT) public double[] scalar; public RapidsNumbersV3() {} public RapidsNumbersV3(double[] ds) { scalar = ds; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/RapidsSchemaV3.java
package water.api.schemas3; import water.Iced; import water.api.API; /** */ public class RapidsSchemaV3<I extends Iced, S extends RapidsSchemaV3<I, S>> extends RequestSchemaV3<I, S> { @API(help="A Rapids AstRoot expression", direction=API.Direction.INPUT, required=true) public String ast; @API(help="Session key", direction=API.Direction.INPUT) public String session_id; @API(help="[DEPRECATED] Key name to assign Frame results", direction=API.Direction.INPUT) public String id; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/RapidsStringV3.java
package water.api.schemas3; import water.Iced; import water.api.API; /** */ public class RapidsStringV3 extends RapidsSchemaV3<Iced,RapidsStringV3> { @API(help="String result", direction=API.Direction.OUTPUT) public String string; public RapidsStringV3() {} public RapidsStringV3(String s) { string = s; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/RapidsStringsV3.java
package water.api.schemas3; import water.Iced; import water.api.API; /** */ public class RapidsStringsV3 extends RapidsSchemaV3<Iced,RapidsStringsV3> { @API(help="String array result", direction=API.Direction.OUTPUT) public String[] string; public RapidsStringsV3() {} public RapidsStringsV3(String[] ss) { string = ss; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/RemoveAllV3.java
package water.api.schemas3; import water.Iced; import water.api.API; public class RemoveAllV3 extends RequestSchemaV3<Iced, RemoveAllV3> { @API(direction = API.Direction.INPUT, help = "Keys of the models to retain", level = API.Level.critical) public KeyV3[] retained_keys; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/RemoveV3.java
package water.api.schemas3; import water.Iced; import water.api.API; public class RemoveV3 extends RequestSchemaV3<Iced, RemoveV3> { @API(help="Object to be removed.") public KeyV3 key; @API(help="If true, removal operation will cascade down the object tree.", direction = API.Direction.INPUT) public boolean cascade; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/RequestSchemaV3.java
package water.api.schemas3; import water.Iced; import water.api.API; import water.api.Schema; /** * Base Schema class for all REST API requests, gathering up common behavior such as the */ public class RequestSchemaV3<I extends Iced, S extends RequestSchemaV3<I, S>> extends SchemaV3<I, S> { @API(help="Comma-separated list of JSON field paths to exclude from the result, used like: " + "\"/3/Frames?_exclude_fields=frames/frame_id/URL,__meta\"", direction=API.Direction.INPUT) public String _exclude_fields = ""; // @API(help="Not yet implemented", direction=API.Direction.INPUT) // public String _include_fields = ""; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/RouteV3.java
package water.api.schemas3; import water.api.API; import water.api.Handler; import water.api.Route; import water.util.PojoUtils; public final class RouteV3 extends SchemaV3<Route, RouteV3> { @API(help="", direction=API.Direction.OUTPUT) public String http_method; @API(help="", direction=API.Direction.OUTPUT) public String url_pattern; @API(help="", direction=API.Direction.OUTPUT) public String summary; @API(help="", direction=API.Direction.OUTPUT) public String api_name; @API(help="", direction=API.Direction.OUTPUT) public String handler_class; @API(help="", direction=API.Direction.OUTPUT) public String handler_method; @API(help="", direction=API.Direction.OUTPUT) public String input_schema; @API(help="", direction=API.Direction.OUTPUT) public String output_schema; // NOTE: Java 7 captures and lets you look up subpatterns by name but won't give you the list of names, so we need this redundant list: @API(help="", direction=API.Direction.OUTPUT) public String[] path_params; // list of params we capture from the url pattern, e.g. for /17/MyComplexObj/(.*)/(.*) @API(help="", direction=API.Direction.OUTPUT) public String markdown; public RouteV3() {} public RouteV3(Route impl) { super(impl); } @Override public RouteV3 fillFromImpl(Route impl) { PojoUtils.copyProperties(this, impl, PojoUtils.FieldNaming.ORIGIN_HAS_UNDERSCORES, new String[] {"url_pattern", "handler_class", "handler_method"} ); this.url_pattern = impl._url; this.handler_class = impl._handler_class.toString(); this.handler_method = impl._handler_method.getName(); this.input_schema = Handler.getHandlerMethodInputSchema(impl._handler_method).getSimpleName(); this.output_schema = Handler.getHandlerMethodOutputSchema(impl._handler_method).getSimpleName(); return this; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/SaveToHiveTableV3.java
package water.api.schemas3; import water.Iced; import water.api.API; import water.api.SaveToHiveTableHandler; public class SaveToHiveTableV3 extends RequestSchemaV3<Iced, SaveToHiveTableV3> { //Input fields @API(help = "H2O Frame ID", required = true) public KeyV3.FrameKeyV3 frame_id; @API(help = "HIVE JDBC URL", required = true) public String jdbc_url; @API(help = "Name of table to save data to.", required = true) public String table_name; @API(help = "HDFS Path to where the table should be stored.") public String table_path; @API(help = "Storage format of the created table.", values = {"CSV", "PARQUET"}) public SaveToHiveTableHandler.HiveFrameSaver.Format format; @API(help = "HDFS Path where to store temporary data.") public String tmp_path; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/SchemaMetadataV3.java
package water.api.schemas3; import water.api.API; import water.api.SchemaMetadata; import water.util.PojoUtils; import java.util.ArrayList; public class SchemaMetadataV3 extends SchemaV3<SchemaMetadata, SchemaMetadataV3> { @API(help="Version number of the Schema.") public int version; /** * The simple schema (class) name, e.g. DeepLearningParametersV2, used in the schema metadata. Must not be * changed after creation (treat as final). */ @API(help="Simple name of the Schema. NOTE: the schema_names form a single namespace.") public String name ; @API(help="[DEPRECATED] This field is always the same as name.", direction=API.Direction.OUTPUT) public String label; /** * The simple schema superclass name, e.g. ModelSchemaV3, used in the schema metadata. Must not be changed after * creation (treat as final). */ @API(help="Simple name of the superclass of the Schema. NOTE: the schema_names form a single namespace.") public String superclass ; @API(help="Simple name of H2O type that this Schema represents. Must not be changed after creation (treat as final).") public String type; @API(help="All the public fields of the schema", direction=API.Direction.OUTPUT) public FieldMetadataV3[] fields; @API(help="Documentation for the schema in Markdown format with GitHub extensions", direction=API.Direction.OUTPUT) String markdown; public SchemaMetadataV3() {} public SchemaMetadataV3(SchemaMetadata impl) { super(impl); } @Override public SchemaMetadata createImpl() { return new SchemaMetadata(); } @Override public SchemaMetadata fillImpl(SchemaMetadata impl) { impl.fields = new ArrayList<>(this.fields.length); int i = 0; for (FieldMetadataV3 s : this.fields) impl.fields.add(s.createImpl()); return impl; } @Override public SchemaMetadataV3 fillFromImpl(SchemaMetadata impl) { PojoUtils.copyProperties(this, impl, PojoUtils.FieldNaming.CONSISTENT, new String[] {"fields"}); this.fields = new FieldMetadataV3[impl.fields.size()]; this.label = impl.name; int i = 0; for (SchemaMetadata.FieldMetadata f : impl.fields) this.fields[i++] = new FieldMetadataV3().fillFromImpl(f); // TODO: version! return this; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/SchemaV3.java
package water.api.schemas3; import water.Iced; import water.api.API; import water.api.Schema; /** * Base Schema class for all v3 REST API objects. * * Any schema which is *NOT* used for requests (i.e. it cannot be an input to a REST API handler) should inherit from * this class. * However if a schema is used for requests, then it should inherit from {@link RequestSchemaV3}, which contains some * additional fields common for all REST API requests. */ public class SchemaV3<I extends Iced, S extends SchemaV3<I,S>> extends Schema<I, S> { @API(help="Metadata on this schema instance, to make it self-describing.", direction=API.Direction.OUTPUT) public Meta __meta; /** * Metadata for a Schema, including the version, name and type. This information is included in all v3 REST API * responses as a field in the Schema so that the payloads are self-describing, and it is also available through * the /Metadata/schemas REST API endpoint for the purposes of REST service discovery. */ public static final class Meta extends Iced { @API(help="Version number of this Schema. Must not be changed after creation (treat as final).", direction=API.Direction.OUTPUT) public int schema_version; @API(help="Simple name of this Schema. NOTE: the schema_names form a single namespace.", direction=API.Direction.OUTPUT) public String schema_name; @API(help="Simple name of H2O type that this Schema represents. Must not be changed after creation (treat as final).", direction=API.Direction.OUTPUT) public String schema_type; /** Default constructor used only for newInstance() in generic reflection-based code. */ public Meta() {} /** Standard constructor which supplies all the fields. The fields should be treated as immutable once set. */ public Meta(int version, String name, String type) { schema_version = version; schema_name = name; schema_type = type; } /** Used during markdown generation. */ public String toString() { return schema_name; } } public SchemaV3() { this(null); } public SchemaV3(I impl) { __meta = new Meta(getSchemaVersion(), getSchemaName(), getSchemaType()); if (impl != null) this.fillFromImpl(impl); } public water.AutoBuffer writeJSON(water.AutoBuffer ab) { // Ugly hack, but sometimes I find that __meta was not initialized by now; which means that constructor was // somehow skipped, which means the object was created in roundabout way and then unsafely cast... Hope we'll // find a proper solution to this issue eventually... if (__meta == null) __meta = new Meta(getSchemaVersion(), getSchemaName(), getSchemaType()); return super.writeJSON(ab); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/SegmentModelsParametersV3.java
package water.api.schemas3; import hex.segments.SegmentModelsBuilder; import water.api.API; public class SegmentModelsParametersV3 extends SchemaV3<SegmentModelsBuilder.SegmentModelsParameters, SegmentModelsParametersV3> { @API(help = "Uniquely identifies the collection of the segment models") public KeyV3.SegmentModelsKeyV3 segment_models_id; @API(help = "Enumeration of all segments for which to build models for") public KeyV3.FrameKeyV3 segments; @API(help = "List of columns to segment-by, models will be built for all segments in the data") public String[] segment_columns; @API(help = "Level of parallelism of bulk model building, it is the maximum number of models each H2O node will be building in parallel") public int parallelism; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/SegmentModelsV3.java
package water.api.schemas3; import water.Iced; import water.api.API; public class SegmentModelsV3 extends SchemaV3<Iced, SegmentModelsV3> { @API(help = "Segment Models id") public KeyV3.SegmentModelsKeyV3 segment_models_id; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/SessionPropertyV3.java
package water.api.schemas3; import water.Iced; import water.api.API; public class SessionPropertyV3 extends RequestSchemaV3<Iced, SessionPropertyV3> { @API(help="Session ID", direction = API.Direction.INOUT) public String session_key; @API(help="Property Key", direction = API.Direction.INOUT) public String key; @API(help="Property Value", direction = API.Direction.INOUT) public String value; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/ShutdownV3.java
package water.api.schemas3; import water.api.ShutdownHandler; public class ShutdownV3 extends RequestSchemaV3<ShutdownHandler.Shutdown,ShutdownV3> { }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/SignificantRulesV3.java
package water.api.schemas3; import water.Iced; import water.api.API; public class SignificantRulesV3 extends RequestSchemaV3<Iced, SignificantRulesV3> { @API(help="Model id of interest", json = false) public KeyV3.ModelKeyV3 model_id; @API(help="The estimated coefficients and language representations (in case it is a rule) for each of the significant baselearners.", direction=API.Direction.OUTPUT) public TwoDimTableV3 significant_rules_table; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/SplitFrameV3.java
package water.api.schemas3; import hex.SplitFrame; import water.api.API; import water.api.schemas3.KeyV3.FrameKeyV3; public class SplitFrameV3 extends SchemaV3<SplitFrame, SplitFrameV3> { @API(help="Job Key") public KeyV3.JobKeyV3 key; @API(help="Dataset") public FrameKeyV3 dataset; @API(help="Split ratios - resulting number of split is ratios.length+1", json=true) public double[] ratios; @API(help="Destination keys for each output frame split.", direction = API.Direction.INOUT) public FrameKeyV3[] destination_frames; public SplitFrameV3() {} public SplitFrameV3(SplitFrame impl) { super(impl); } @Override public SplitFrame createImpl() { return new SplitFrame(); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/SteamMetricsV3.java
package water.api.schemas3; import water.api.API; public class SteamMetricsV3 extends RequestSchemaV3<SteamMetricsV3.SteamMetrics, SteamMetricsV3> { public static class SteamMetrics extends water.Iced<SteamMetrics> { public SteamMetrics() {} } @API(help="Steam metrics API version", direction = API.Direction.OUTPUT) public long version; @API(help="Number of milliseconds that the cluster has been idle", direction = API.Direction.OUTPUT) public long idle_millis; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/StringPairV3.java
package water.api.schemas3; import hex.StringPair; import water.api.API; import water.api.Schema; public class StringPairV3 extends SchemaV3<StringPair, StringPairV3> implements Schema.AutoParseable { @API(help = "Value A") String a; @API(help = "Value B") String b; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/TabulateV3.java
package water.api.schemas3; import water.api.API; import water.api.schemas3.KeyV3.FrameKeyV3; import water.util.Tabulate; public class TabulateV3 extends SchemaV3<Tabulate, TabulateV3> { // INPUT @API(help="Dataset", required = true) public FrameKeyV3 dataset; @API(help="Predictor", required = true, level = API.Level.critical, is_member_of_frames = {"dataset"}, is_mutually_exclusive_with = {"col_y"}, direction = API.Direction.INOUT) public FrameV3.ColSpecifierV3 predictor; @API(help="Response", required = true, level = API.Level.critical, is_member_of_frames = {"dataset"}, is_mutually_exclusive_with = {"col_x"}, direction = API.Direction.INOUT) public FrameV3.ColSpecifierV3 response; @API(help="Observation weights (optional)", required = false, level = API.Level.critical, is_member_of_frames = {"dataset"}, is_mutually_exclusive_with = {"col_x"}, direction = API.Direction.INOUT) public FrameV3.ColSpecifierV3 weight; @API(help="Number of bins for predictor column") public int nbins_predictor; @API(help="Number of bins for response column") public int nbins_response; // OUTPUT @API(help="Counts table", direction = API.Direction.OUTPUT) public TwoDimTableV3 count_table; @API(help="Response table", direction = API.Direction.OUTPUT) public TwoDimTableV3 response_table; public TabulateV3() {} public TabulateV3(Tabulate impl) { super(impl); } @Override public TabulateV3 fillFromImpl(Tabulate impl) { super.fillFromImpl(impl); return this; } @Override public Tabulate createImpl() { return new Tabulate(); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/TimelineV3.java
package water.api.schemas3; import water.*; import water.api.API; import water.api.TimelineHandler.Timeline; import water.init.TimelineSnapshot; import java.text.SimpleDateFormat; import java.util.ArrayList; import java.util.Date; /** * Display of a Timeline */ public class TimelineV3 extends RequestSchemaV3<Timeline,TimelineV3> { // This schema has no input params @API(help="Current time in millis.", direction=API.Direction.OUTPUT) public long now; @API(help="This node", direction=API.Direction.OUTPUT) public String self; @API(help="recorded timeline events", direction=API.Direction.OUTPUT) public EventV3[] events; public static class EventV3<I, S extends EventV3<I, S>> extends SchemaV3<Iced, S> { @API(help="Time when the event was recorded. Format is hh:mm:ss:ms") public final String date; @API(help="Time in nanos") public final long nanos; enum EventType {unknown, heartbeat, network_msg, io} @API(help="type of recorded event", values = {"unknown", "heartbeat", "network_msg", "io"}) public final EventType type; @SuppressWarnings("unused") public EventV3() { date = null; nanos = -1; type = EventType.unknown; } private EventV3(EventType type, long millis, long nanos){ this.type = type; this.date = new SimpleDateFormat("HH:mm:ss:SSS").format(new Date(millis)); this.nanos = nanos; } protected String who() { throw H2O.unimpl(); }; protected String ioType() { throw H2O.unimpl(); }; protected String event() { throw H2O.unimpl(); }; public String bytes() { throw H2O.unimpl(); }; } // Event public static class HeartBeatEvent extends EventV3<Iced, HeartBeatEvent> { @API(help = "number of sent heartbeats") final int sends; @API(help = "number of received heartbeats") final int recvs; public HeartBeatEvent() { super(); sends = -1; recvs = -1; } private HeartBeatEvent(int sends, int recvs, long lastMs, long lastNs){ super(EventType.heartbeat,lastMs,lastNs); this.sends = sends; this.recvs = recvs; } @Override protected String who() { return "many -> many";} @Override protected String ioType() {return "UDP";} @Override protected String event() {return "heartbeat";} @Override public String bytes() {return sends + " sent " + ", " + recvs + " received";} @Override public String toString() { return "HeartBeat(" + sends + " sends, " + recvs + " receives)"; } } // HeartBeatEvent public static class NetworkEvent extends EventV3<Iced, NetworkEvent> { @API(help="Boolean flag distinguishing between sends (true) and receives(false)") public final boolean is_send; @API(help="network protocol (UDP/TCP)") public final String protocol; @API(help="UDP type (exec,ack, ackack,...") public final String msg_type; // udp @API(help="Sending node") public final String from; @API(help="Receiving node") public final String to; @API(help="Pretty print of the first few bytes of the msg payload. Contains class name for tasks.") public final String data; public NetworkEvent() { super(); is_send = false; protocol = "unknown"; msg_type = "unknown"; from = "unknown"; to = "unknown"; data = "unknown"; } private NetworkEvent(long ms, long ns, boolean is_send, String protocol, String msg_type, String from, String to, String data){ super(EventType.network_msg,ms,ns); this.is_send = is_send; this.protocol = protocol; this.msg_type = msg_type; this.from = from; this.to = to; this.data = data; } @Override protected String who() { return from + " -> " + to;} @Override protected String ioType() {return protocol;} @Override protected String event() {return msg_type;} @Override public String bytes() {return data;} @Override public String toString() { return "NetworkMsg(" + from + " -> " + to + ", protocol = '" + protocol + "', data = '" + data + "')"; } } // NetworkEvent public static class IOEvent extends EventV3<Iced, IOEvent> { @API(help="flavor of the recorded io (ice/hdfs/...)") public final String io_flavor; @API(help="node where this io event happened") public final String node; @API(help="data info") public final String data; public IOEvent() { this(-1, -1, "unknown", "unknown", "unknown"); } private IOEvent(long ms, long ns, String node, String io_flavor, String data){ super(EventType.io,ms,ns); this.io_flavor = io_flavor; this.node = node; this.data = data; } @Override protected String who(){return node;} @Override protected String ioType() {return io_flavor;} @Override protected String event() {return "i_o";} @Override public String bytes() { return data;} @Override public String toString() { return "I_O('" + io_flavor + "')"; } } // IOEvent @Override public TimelineV3 fillFromImpl(Timeline timeline) { ArrayList<EventV3> outputEvents = new ArrayList<>(); ArrayList<TimelineSnapshot.Event> heartbeats = new ArrayList(); H2O cloud = TimeLine.getCLOUD(); if (null != timeline.snapshot) { for (TimelineSnapshot.Event event : timeline.snapshot) { H2ONode h2o = cloud.members()[event._nodeId]; // The event type. First get payload. UDP.udp msgType = event.udpType(); // Accumulate repeated heartbeats if (msgType == UDP.udp.heartbeat) { heartbeats.add(event); continue; } // Now dump out accumulated heartbeats if (!heartbeats.isEmpty()) { long firstMs = heartbeats.get(0).ms(); long lastMs = heartbeats.get(heartbeats.size() - 1).ms(); int totalSends = 0; int totalRecvs = 0; int totalDrops = 0; int[] sends = new int[cloud.size()]; int[] recvs = new int[cloud.size()]; for (TimelineSnapshot.Event h : heartbeats) { if (h.isSend()) { ++totalSends; ++sends[h._nodeId]; } else if (h.isDropped()) { ++totalDrops; } else { ++totalRecvs; ++recvs[h._nodeId]; } } heartbeats.clear(); outputEvents.add(new HeartBeatEvent(totalSends, totalRecvs, firstMs, lastMs)); } long ms = event.ms(); long ns = event.ns(); if (msgType == UDP.udp.i_o) { // handle io event outputEvents.add(new IOEvent(ms, ns, event.recoH2O().toString(), event.ioflavor(), UDP.printx16(event.dataLo(), event.dataHi()))); } else { // network msg String from, to; if (event.isSend()) { from = h2o.toString(); to = event.packH2O() == null ? "multicast" : event.packH2O().toString(); } else { from = event.packH2O().toString(); to = h2o.toString(); } outputEvents.add(new NetworkEvent(ms, ns, event.isSend(), event.isTCP() ? "TCP" : "UDP", msgType.toString(), from, to, UDP.printx16(event.dataLo(), event.dataHi()))); } } } // if timeline.snapshot events = outputEvents.toArray(new EventV3[null == outputEvents ? 0 : outputEvents.size()]); return this; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/TwoDimTableV3.java
package water.api.schemas3; import water.AutoBuffer; import water.H2O; import water.Iced; import water.IcedWrapper; import water.api.API; import water.util.TwoDimTable; /** * Client-facing Schema of a TwoDimTable * Notes: * 1) We embed the rowHeaders into the table, extending it by 1 column * 2) We store all the data in column-major order * 3) We store all the data in String format * */ public class TwoDimTableV3 extends SchemaV3<TwoDimTable, TwoDimTableV3> { public static class ColumnSpecsBase extends SchemaV3<Iced, ColumnSpecsBase> { @API(help="Column Name", direction=API.Direction.OUTPUT) public String name; // allow reset of col header names so that I can undo the pythonize for GLM multinomial coeff names @API(help="Column Type", direction=API.Direction.OUTPUT) String type; @API(help="Column Format (printf)", direction=API.Direction.OUTPUT) String format; @API(help="Column Description", direction=API.Direction.OUTPUT) String description; } @API(help="Table Name", direction=API.Direction.OUTPUT) public String name; @API(help="Table Description", direction=API.Direction.OUTPUT) public String description; @API(help="Column Specification", direction=API.Direction.OUTPUT) public ColumnSpecsBase[] columns; @API(help="Number of Rows", direction=API.Direction.OUTPUT) public int rowcount; @API(help="Table Data (col-major)", direction=API.Direction.OUTPUT) public IcedWrapper[][] data; public TwoDimTableV3() {} public TwoDimTableV3(TwoDimTable impl) { super(impl); } /** * Fill a TwoDimTable Schema from a TwoDimTable * @param t TwoDimTable * @return TwoDimTableSchema */ @Override public TwoDimTableV3 fillFromImpl(TwoDimTable t) { name = t.getTableHeader(); description = t.getTableDescription(); final int rows = t.getRowDim(); rowcount = rows; boolean have_row_header_cols = t.getColHeaderForRowHeaders() != null; for (int r=0; r<rows; ++r) { if (!have_row_header_cols) break; have_row_header_cols &= t.getRowHeaders()[r] != null; } if (have_row_header_cols) { final int cols = t.getColDim()+1; columns = new ColumnSpecsBase[cols]; columns[0] = new ColumnSpecsBase(); columns[0].name = pythonify(t.getColHeaderForRowHeaders()); columns[0].type = "string"; columns[0].format = "%s"; columns[0].description = t.getColHeaderForRowHeaders(); for (int c = 1; c < cols; ++c) { columns[c] = new ColumnSpecsBase(); columns[c].name = pythonify(t.getColHeaders()[c - 1]); columns[c].type = t.getColTypes()[c - 1]; columns[c].format = t.getColFormats()[c - 1]; columns[c].description = t.getColHeaders()[c - 1]; } data = new IcedWrapper[cols][rows]; data[0] = new IcedWrapper[t.getRowDim()]; for (int r = 0; r < t.getRowDim(); ++r) { data[0][r] = new IcedWrapper(t.getRowHeaders()[r]); } IcedWrapper[][] cellValues = t.getCellValues(); for (int c = 1; c < cols; ++c) { data[c] = new IcedWrapper[rows]; for (int r = 0; r < rows; ++r) { data[c][r] = cellValues[r][c - 1]; } } } else { final int cols = t.getColDim(); columns = new ColumnSpecsBase[cols]; for (int c = 0; c < cols; ++c) { columns[c] = new ColumnSpecsBase(); columns[c].name = pythonify(t.getColHeaders()[c]); columns[c].type = t.getColTypes()[c]; columns[c].format = t.getColFormats()[c]; columns[c].description = t.getColHeaders()[c]; } data = new IcedWrapper[cols][rows]; IcedWrapper[][] cellValues = t.getCellValues(); for (int c = 0; c < cols; ++c) { data[c] = new IcedWrapper[rows]; for (int r = 0; r < rows; ++r) { data[c][r] = cellValues[r][c]; } } } return this; } /** * Turn a description such as "Avg. Training MSE" into a JSON-usable field name "avg_training_mse" * @param n * @return */ private String pythonify(String n) { if (n == null || name.toLowerCase().contains("confusion")) return n; StringBuilder sb = new StringBuilder(); String [] modified = n.split("[\\s_]+"); for (int i=0; i<modified.length; ++i) { if (i!=0) sb.append("_"); String s = modified[i]; // if (!s.matches("^[A-Z]{2,3}$")) { sb.append(s.toLowerCase()); //everything goes lowercase // } else { // sb.append(s); // } } String newString = sb.toString().replaceAll("[^\\w]", ""); // if (!newString.equals(name)) { // Log.warn("Turning column description into field name: " + name + " --> " + newString); // } return newString; } /** * Fill a TwoDimTable from this Schema * @param impl * @return */ public TwoDimTable fillImpl(TwoDimTable impl) { final int rows = data[0].length; assert(rows == rowcount); final int cols = data.length+1; String tableHeader = name; String tableDescription = description; String colHeaderForRowHeaders = columns[0].name; String[] rowHeaders = new String[rows]; for (int r=0; r<rows; ++r) { rowHeaders[r] = (String)data[0][r].get(); } String[] colHeaders = new String[cols]; colHeaders[0] = ""; for (int c=1; c<cols; ++c) { colHeaders[c] = columns[c].description; } String[] colTypes = new String[cols]; colTypes[0] = ""; for (int c=1; c<cols; ++c) { colTypes[c] = columns[c].type; } String[] colFormats = new String[cols]; colFormats[0] = "%s"; for (int c=1; c<cols; ++c) { colFormats[c] = columns[c].format; } String[][] strCellValues = new String[rows][cols]; double[][] dblCellValues = new double[rows][cols]; for (int r=0; r<data[0].length; ++r) { for (int c=0; c<data.length; ++c) { try { if (columns[c].format.equals("string")) { // switch(String) is not java1.6 compliant! strCellValues[r][c] = (String)data[c][r].get(); } else if (columns[c].format.equals("double")) { dblCellValues[r][c] = (Double)data[c][r].get(); } else if (columns[c].format.equals("float")) { dblCellValues[r][c] = (Float)data[c][r].get(); } else if (columns[c].format.equals("int")) { dblCellValues[r][c] = (Integer)data[c][r].get(); } else if (columns[c].format.equals("long")) { dblCellValues[r][c] = (Long)data[c][r].get(); } else throw H2O.fail(); } catch (ClassCastException e) { throw new RuntimeException(e); } } } return new TwoDimTable(tableHeader, tableDescription, rowHeaders, colHeaders, colTypes, colFormats, colHeaderForRowHeaders, strCellValues, dblCellValues); } public final AutoBuffer writeJSON_impl(AutoBuffer ab) { ab.putJSONStr("name",name); ab.put1(','); ab.putJSONStr("description",description); ab.put1(','); ab.putJSONStr("columns").put1(':'); ab.put1('['); if( columns!=null ) { for (int i = 0; i < columns.length; ++i) { columns[i].writeJSON(ab); if (i < columns.length - 1) ab.put1(','); } } ab.put1(']'); ab.put1(','); ab.putJSON4("rowcount", rowcount); ab.put1(','); ab.putJSONStr("data").put1(':'); ab.put1('['); if( data!=null ) { for (int i = 0; i < data.length; ++i) { ab.put1('['); for (int j = 0; j < data[i].length; ++j) { if (data[i][j] == null || data[i][j].get() == null) { ab.putJNULL(); } else { data[i][j].writeUnwrappedJSON(ab); } if (j < data[i].length - 1) ab.put1(','); } ab.put1(']'); if (i < data.length - 1) ab.put1(','); } } ab.put1(']'); return ab; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/TypeaheadV3.java
package water.api.schemas3; import water.Iced; import water.api.API; public class TypeaheadV3 extends RequestSchemaV3<Iced,TypeaheadV3> { // Input fields @API(help="training_frame", required=true) public String src; @API(help="limit") public int limit; // Output fields @API(help="matches", direction=API.Direction.OUTPUT) public String matches[]; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/UnlockKeysV3.java
package water.api.schemas3; import water.Iced; public class UnlockKeysV3 extends RequestSchemaV3<Iced, UnlockKeysV3> { }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/ValidationMessageV3.java
package water.api.schemas3; import hex.ModelBuilder; import hex.ModelBuilder.ValidationMessage; import water.api.API; import water.util.Log; import water.util.PojoUtils; import java.util.HashMap; import java.util.Map; /** * Model builder parameter validation message schema. */ public final class ValidationMessageV3 extends SchemaV3<ValidationMessage, ValidationMessageV3> { @API(help = "Type of validation message (ERROR, WARN, INFO, HIDE)", direction = API.Direction.OUTPUT) public String message_type; @API(help = "Field to which the message applies", direction = API.Direction.OUTPUT) public String field_name; @API(help = "Message text", direction = API.Direction.OUTPUT) public String message; /** * Map impl field names in the validation messages to schema field names, * called <i>after</i> behavior of stripping leading _ characters. */ public static void mapValidationMessageFieldNames(ValidationMessageV3[] validation_messages, String[] from, String[] to) { if (from == null && to == null) return; if (from == null || to == null) throw new IllegalArgumentException("Bad parameter name translation arrays; one is null and the other isn't."); Map<String, String> translations = new HashMap<>(); for (int i = 0; i < from.length; i++) { translations.put(from[i], to[i]); } for (ValidationMessageV3 vm : validation_messages) { if (null == vm) { Log.err("Null ValidationMessageV3 for ModelBuilderSchema."); continue; } if (null == vm.field_name) { Log.err("Null field_name: " + vm); continue; } if (translations.containsKey(vm.field_name)) vm.field_name = translations.get(vm.field_name); } } public ValidationMessage createImpl() { return new ModelBuilder.ValidationMessage(Log.valueOf(message_type), field_name, message); } // Version&Schema-specific filling from the implementation object public ValidationMessageV3 fillFromImpl(ValidationMessage vm) { PojoUtils.copyProperties(this, vm, PojoUtils.FieldNaming.ORIGIN_HAS_UNDERSCORES); this.message_type = Log.LVLS[vm.log_level()]; // field name changed if (this.field_name != null) { if (this.field_name.startsWith("_")) this.field_name = this.field_name.substring(1); else Log.warn("Expected all ValidationMessage field_name values to have leading underscores; ignoring: " + field_name); } return this; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/VarImpV3.java
package water.api.schemas3; import hex.VarImp; import water.api.API; public class VarImpV3 extends SchemaV3<VarImp,VarImpV3> { @API(help="Variable importance of individual variables", direction=API.Direction.OUTPUT) public float[] varimp; @API(help="Names of variables", direction=API.Direction.OUTPUT) protected String[] names; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/WaterMeterCpuTicksV3.java
package water.api.schemas3; import water.api.API; import water.util.PojoUtils; import water.util.WaterMeterCpuTicks; public class WaterMeterCpuTicksV3 extends RequestSchemaV3<WaterMeterCpuTicks, WaterMeterCpuTicksV3> { @API(help="Index of node to query ticks for (0-based)", required = true, direction = API.Direction.INPUT) public int nodeidx; @API(help="array of tick counts per core", direction = API.Direction.OUTPUT) public long[][] cpu_ticks; // Version&Schema-specific filling into the implementation object public WaterMeterCpuTicks createImpl() { WaterMeterCpuTicks obj = new WaterMeterCpuTicks(); PojoUtils.copyProperties(obj, this, PojoUtils.FieldNaming.CONSISTENT); return obj; } // Version&Schema-specific filling from the implementation object public WaterMeterCpuTicksV3 fillFromImpl(WaterMeterCpuTicks i) { PojoUtils.copyProperties(this, i, PojoUtils.FieldNaming.CONSISTENT); return this; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas3/WaterMeterIoV3.java
package water.api.schemas3; import water.api.API; import water.util.PojoUtils; import water.util.WaterMeterIo; public class WaterMeterIoV3 extends RequestSchemaV3<WaterMeterIo, WaterMeterIoV3> { @API(help="Index of node to query ticks for (0-based)", direction = API.Direction.INPUT) public int nodeidx; @API(help="array of IO info", direction = API.Direction.OUTPUT) public WaterMeterIo.IoStatsEntry persist_stats[]; // Version&Schema-specific filling into the implementation object public WaterMeterIo createImpl() { WaterMeterIo obj = new WaterMeterIo(); PojoUtils.copyProperties(obj, this, PojoUtils.FieldNaming.CONSISTENT); return obj; } // Version&Schema-specific filling from the implementation object public WaterMeterIoV3 fillFromImpl(WaterMeterIo i) { PojoUtils.copyProperties(this, i, PojoUtils.FieldNaming.CONSISTENT); return this; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/EndpointV4.java
package water.api.schemas4; import water.api.API; import water.api.Handler; import water.api.Route; /** * */ public class EndpointV4 extends OutputSchemaV4<Route, EndpointV4> { @API(help="Method+Url of the request; variable parts are enclosed in curly braces. For example: " + "/4/schemas/{schema_name}") public String url; @API(help="Short description of the functionality provided by the endpoint.") public String description; @API(help="Unique name of the endpoint. These names can be used to look up the endpoint's info via " + "GET /4/endpoints/{name}.") public String name; // TODO: more explanation -- how input object corresponds to the actual request @API(help="Input schema.") public String input_schema; @API(help="Schema for the result returned by the endpoint.") public String output_schema; @Override public EndpointV4 fillFromImpl(Route route) { url = route._http_method + " " + route._url; description = route._summary; name = route._api_name; input_schema = "/4/schemas/" + Handler.getHandlerMethodInputSchema(route._handler_method).getSimpleName(); output_schema = "/4/schemas/" + Handler.getHandlerMethodOutputSchema(route._handler_method).getSimpleName(); return this; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/EndpointsListV4.java
package water.api.schemas4; import water.Iced; import water.api.API; /** * List of endpoints, returned by GET /4/endpoints */ public class EndpointsListV4 extends OutputSchemaV4<Iced, EndpointsListV4> { @API(help="List of endpoints in H2O REST API (v4).") public EndpointV4[] endpoints; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/InputSchemaV4.java
package water.api.schemas4; import water.Iced; import water.api.API; import water.api.Schema; /** * Base Schema class for all v4 REST API requests. It provides common _schema field, as well as . */ public class InputSchemaV4<I extends Iced, S extends InputSchemaV4<I,S>> extends Schema<I,S> { @API(help="Filter on the set of output fields: if you set _fields=\"foo,bar,baz\", then only those fields will be " + "included in the output; or you can specify _fields=\"-goo,gee\" to include all fields except goo and gee. If " + "the result contains nested data structures, then you can refer to the fields within those structures as well. " + "For example if you specify _fields=\"foo(oof),bar(-rab)\", then only fields foo and bar will be included, and " + "within foo there will be only field oof, whereas within bar all fields except rab will be reported.") public String _fields; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/ListRequestV4.java
package water.api.schemas4; import water.Iced; /** * Common input schema class for endpoints that request collections of objects. For example, * GET /4/schemas * GET /4/frames * GET /4/models * etc. * This class is a placeholder right now, but eventually it can host functionality such as filtering/sorting the * results, providing cursor capabilities, etc. * TODO add cursor fields {limit} and {offset} */ public class ListRequestV4 extends OutputSchemaV4<Iced, ListRequestV4> { }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/ModelInfoV4.java
package water.api.schemas4; import hex.ModelBuilder; import water.api.API; /** * Lightweight information profile about each model. */ public class ModelInfoV4 extends OutputSchemaV4<ModelBuilder, ModelInfoV4> { @API(help="Algorithm name, such as 'gbm', 'deeplearning', etc.") public String algo; @API(help="Development status of the algorithm: alpha, beta, or stable.") public String maturity; @API(help="Does the model support generation of POJOs?") public boolean have_pojo; @API(help="Does the model support generation of MOJOs?") public boolean have_mojo; @API(help="Mojo version number for this algorithm.") public String mojo_version; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/ModelsInfoV4.java
package water.api.schemas4; import water.Iced; import water.api.API; /** * List of models, returned by GET /4/modelsinfo */ public class ModelsInfoV4 extends OutputSchemaV4<Iced, ModelsInfoV4> { @API(help="Generic information about each model supported in H2O.") public ModelInfoV4[] models; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/OutputSchemaV4.java
package water.api.schemas4; import water.Iced; import water.api.API; import water.api.Schema; /** * Base output Schema class for all v4 REST API requests. It provides common __schema field that identifies the * schema in the output. */ public class OutputSchemaV4<I extends Iced, S extends OutputSchemaV4<I,S>> extends Schema<I,S> { @API(help="Url describing the schema of the current object.") public String __schema; public OutputSchemaV4() { __schema = "/4/schemas/" + this.getSchemaName(); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/SessionIdV4.java
package water.api.schemas4; import water.Iced; import water.api.API; public class SessionIdV4 extends OutputSchemaV4<Iced, SessionIdV4> { @API(help="Session ID") public String session_key; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/input/CreateFrameOriginalIV4.java
package water.api.schemas4.input; import hex.createframe.recipes.OriginalCreateFrameRecipe; import water.api.API; import water.api.schemas3.KeyV3; import water.api.schemas4.InputSchemaV4; /** * Input schema for `POST /4/frames/$original` endpoint. */ public class CreateFrameOriginalIV4 extends InputSchemaV4<OriginalCreateFrameRecipe, CreateFrameOriginalIV4> { @API(help="destination key") public KeyV3.FrameKeyV3 dest; @API(help = "Number of rows") public int rows; @API(help = "Number of data columns (in addition to the first response column)") public int cols; @API(help = "Random number seed that determines the random values") public long seed; @API(help = "Whether frame should be randomized") public boolean randomize; @API(help = "Constant value (for randomize=false)") public long value; @API(help = "Range for real variables (-range ... range)") public double real_range; @API(help = "Fraction of categorical columns (for randomize=true)") public double categorical_fraction; @API(help = "Factor levels for categorical variables") public int factors; @API(help = "Fraction of integer columns (for randomize=true)") public double integer_fraction; @API(help = "Range for integer variables (-range ... range)") public int integer_range; @API(help = "Fraction of binary columns (for randomize=true)") public double binary_fraction; @API(help = "Fraction of 1's in binary columns") public double binary_ones_fraction; @API(help = "Fraction of date/time columns (for randomize=true)") public double time_fraction; @API(help = "Fraction of string columns (for randomize=true)") public double string_fraction; @API(help = "Fraction of missing values") public double missing_fraction; @API(help = "Whether an additional response column should be generated") public boolean has_response; @API(help = "Number of factor levels of the first column (1=real, 2=binomial, N=multinomial)") public int response_factors; @API(help = "For real-valued response variable: Whether the response should be positive only.") public boolean positive_response; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/input/CreateFrameSimpleIV4.java
package water.api.schemas4.input; import hex.createframe.recipes.SimpleCreateFrameRecipe; import water.api.API; import water.api.schemas3.KeyV3; import water.api.schemas4.InputSchemaV4; /** * Input schema for `POST /4/frames/$simple` endpoint. */ public class CreateFrameSimpleIV4 extends InputSchemaV4<SimpleCreateFrameRecipe, CreateFrameSimpleIV4> { @API(help = "Id for the frame to be created.") public KeyV3.FrameKeyV3 dest; @API(help = "Random number seed that determines the random values.") public long seed; @API(help = "Number of rows.") public int nrows; @API(help = "Number of real-valued columns. Values in these columns will be uniformly distributed between " + "real_lb and real_ub.") public int ncols_real; @API(help = "Number of integer columns.") public int ncols_int; @API(help = "Number of enum (categorical) columns.") public int ncols_enum; @API(help = "Number of boolean (binary) columns.") public int ncols_bool; @API(help = "Number of string columns.") public int ncols_str; @API(help = "Number of time columns.") public int ncols_time; @API(help = "Lower bound for the range of the real-valued columns.") public double real_lb; @API(help = "Upper bound for the range of the real-valued columns.") public double real_ub; @API(help = "Lower bound for the range of integer columns.") public int int_lb; @API(help = "Upper bound for the range of integer columns.") public int int_ub; @API(help = "Number of levels (categories) for the enum columns.") public int enum_nlevels; @API(help = "Fraction of ones in each boolean (binary) column.") public double bool_p; @API(help = "Lower bound for the range of time columns (in ms since the epoch).") public long time_lb; @API(help = "Upper bound for the range of time columns (in ms since the epoch).") public long time_ub; @API(help = "Length of generated strings in string columns.") public int str_length; @API(help = "Fraction of missing values.") public double missing_fraction; @API(help = "Type of the response column to add.", values = {"none", "real", "int", "bool", "enum", "time"}) public SimpleCreateFrameRecipe.ResponseType response_type; @API(help = "Lower bound for the response variable (real/int/time types).") public double response_lb; @API(help = "Upper bound for the response variable (real/int/time types).") public double response_ub; @API(help = "Frequency of 1s for the bool (binary) response column.") public double response_p; @API(help = "Number of categorical levels for the enum response column.") public int response_nlevels; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/input/JobIV4.java
package water.api.schemas4.input; import water.Iced; import water.api.API; import water.api.schemas4.InputSchemaV4; /** * Input schema for the {@code "GET /4/jobs/{job_id}"} endpoint. */ public class JobIV4 extends InputSchemaV4<Iced, JobIV4> { @API(help="Id of the job to fetch.") public String job_id; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas4/output/JobV4.java
package water.api.schemas4.output; import water.Job; import water.TypeMap; import water.api.API; import water.api.schemas4.OutputSchemaV4; import java.io.PrintWriter; import java.io.StringWriter; /** Schema for a single Job. */ public class JobV4 extends OutputSchemaV4<Job<?>, JobV4> { @API(help="Job id") public String job_id; @API(help="Job status", values={"RUNNING", "DONE", "STOPPING", "CANCELLED", "FAILED"}) public Status status; @API(help="Current progress, a number going from 0 to 1") public float progress; @API(help="Current progress status description") public String progress_msg; @API(help="Start time") public long start_time; @API(help="Runtime in milliseconds") public long duration; @API(help="Id of the target object (being created by this Job)") public String target_id; @API(help="Type of the target: Frame, Model, etc.") public String target_type; @API(help="Exception message, if an exception occurred") public String exception; @API(help="Stacktrace") public String stacktrace; public enum Status { RUNNING, DONE, STOPPING, CANCELLED, FAILED } @Override public JobV4 fillFromImpl(Job<?> job) { if (job == null) return this; job_id = job._key.toString(); progress = job.progress(); progress_msg = job.progress_msg(); duration = job.msec(); if (job.isRunning()) { status = job.stop_requested()? Status.STOPPING : Status.RUNNING; } else { status = job.stop_requested()? Status.CANCELLED : Status.DONE; } Throwable ex = job.ex(); if (ex != null) { status = Status.FAILED; exception = ex.toString(); StringWriter sw = new StringWriter(); ex.printStackTrace(new PrintWriter(sw)); stacktrace = sw.toString(); } target_id = job._result == null || !job.readyForView()? null : job._result.toString(); target_type = TypeMap.theFreezable(job._typeid).getClass().getSimpleName(); return this; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas99/AssemblyV99.java
package water.api.schemas99; import water.Iced; import water.api.API; import water.api.schemas3.KeyV3; import water.api.schemas3.RequestSchemaV3; /** FIXME: comments please! */ public class AssemblyV99 extends RequestSchemaV3<Iced, AssemblyV99> { // input fields @API(help="A list of steps describing the assembly line.") public String[] steps; @API(help="Input Frame for the assembly.") public KeyV3.FrameKeyV3 frame; @API(help="The name of the file (and generated class in case of pojo)") public String file_name; @API(help="The key of the Assembly object to retrieve from the DKV.") public String assembly_id; //output @API(help="Output of the assembly line.", direction=API.Direction.OUTPUT) public KeyV3.FrameKeyV3 result; @API(help="A Key to the fit Assembly data structure", direction=API.Direction.OUTPUT) public KeyV3.AssemblyKeyV3 assembly; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas99/GridsV99.java
package water.api.schemas99; import hex.schemas.GridSchemaV99; import water.api.API; import water.api.schemas3.SchemaV3; import water.api.Grids; public class GridsV99 extends SchemaV3<Grids, GridsV99> { @API(help="Grids", direction=API.Direction.OUTPUT) public GridSchemaV99[] grids; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api
java-sources/ai/h2o/h2o-core/3.46.0.7/water/api/schemas99/RapidsV99.java
package water.api.schemas99; import water.Iced; import water.api.API; import water.api.schemas3.RequestSchemaV3; import water.api.schemas3.KeyV3; public class RapidsV99 extends RequestSchemaV3<Iced, RapidsV99> { // Input fields @API(help="An Abstract Syntax Tree.", direction=API.Direction.INPUT) public String ast; // Output. Only one of these 5 results is returned; the rest are null - and // this is how the caller tells about which result is valid. @API(help="Parsing error, if any", direction=API.Direction.OUTPUT) public String error; @API(help="Scalar result", direction=API.Direction.OUTPUT) public double scalar; @API(help="Function result", direction=API.Direction.OUTPUT) public String funstr; @API(help="String result", direction=API.Direction.OUTPUT) public String string; @API(help="Result key", direction=API.Direction.OUTPUT) public KeyV3.FrameKeyV3 key; @API(help="Rows in Frame result", direction=API.Direction.OUTPUT) public long num_rows; @API(help="Columns in Frame result", direction=API.Direction.OUTPUT) public int num_cols; }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/codegen/CodeGenerator.java
package water.codegen; import water.exceptions.JCodeSB; /** * Interface for code generator. */ public interface CodeGenerator { /** Generate code to given output. * * @param out code generation output. */ void generate(JCodeSB out); }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/codegen/CodeGeneratorPipeline.java
package water.codegen; import java.util.ArrayList; import water.exceptions.JCodeSB; /** * A simple code generation pipeline. * * It composes code generators and allows for their execution * later. */ public class CodeGeneratorPipeline extends ArrayList<CodeGenerator> implements CodeGenerator { @Override public void generate(JCodeSB out) { for (CodeGenerator codeGen : this) { codeGen.generate(out); } } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OAbstractRuntimeException.java
package water.exceptions; import water.util.HttpResponseStatus; import water.H2OError; import water.util.IcedHashMapGeneric; /** * RuntimeException which results in a http 400 error by default, and serves as a base class for other error types. * Note that the HTTP error status can be overridden when converting the exception to {@link H2OError} * when overriding {@link #toH2OError(String)}. * * NOTE: don't use this directly; use more specific types. */ abstract public class H2OAbstractRuntimeException extends RuntimeException { protected int HTTP_RESPONSE_CODE() { return HttpResponseStatus.BAD_REQUEST.getCode(); } public long timestamp; public String dev_message; public IcedHashMapGeneric.IcedHashMapStringObject values; public H2OAbstractRuntimeException(String message, String dev_message, IcedHashMapGeneric.IcedHashMapStringObject values) { super(message); this.timestamp = System.currentTimeMillis(); this.dev_message = dev_message; this.values = values; } public H2OAbstractRuntimeException(String msg, String dev_msg) { this(msg, dev_msg, new IcedHashMapGeneric.IcedHashMapStringObject()); } public H2OError toH2OError() { return toH2OError(null); } public H2OError toH2OError(String error_url) { return new H2OError(timestamp, error_url, getMessage(), dev_message, HTTP_RESPONSE_CODE(), values, this); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OCategoricalLevelNotFoundArgumentException.java
package water.exceptions; import water.util.IcedHashMap; import water.util.IcedHashMapGeneric; /** * Exception signalling that an categorical level was not found. */ public class H2OCategoricalLevelNotFoundArgumentException extends H2ONotFoundArgumentException { public H2OCategoricalLevelNotFoundArgumentException(String argument, String categorical_level, String frame_name, String column_name) { super("Categorical level: " + categorical_level + " not found in column_name: " + column_name + " in frame: " + frame_name + " from argument: " + argument + ": " + argument.toString(), "Categorical level: " + categorical_level + " not found in column_name: " + column_name + " in frame: " + frame_name + " from argument: " + argument + ": " + argument.toString()); this.values = new IcedHashMapGeneric.IcedHashMapStringObject(); this.values.put("argument", argument); this.values.put("categorical_level", categorical_level); this.values.put("frame_name", frame_name); this.values.put("column_name", column_name); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OColumnNotFoundArgumentException.java
package water.exceptions; import water.fvec.Frame; import water.util.IcedHashMap; import water.util.IcedHashMapGeneric; /** * Exception signalling that a Vec was not found. * <p> * If the Vec name came from an argument, especially from an API parameter, use * {@code H2OColumnNotFoundArgumentException(String argument, Frame frame, String column_name)} or * {@code H2OColumnNotFoundArgumentException(String argument, String frame_name, String column_name)}, * which let you specify the argument name. If not, use * {@code H2OColumnNotFoundArgumentException(Frame frame, String column_name)} or * {@code H2OColumnNotFoundArgumentException(String frame_name, String column_name)}. */ public class H2OColumnNotFoundArgumentException extends H2ONotFoundArgumentException { public H2OColumnNotFoundArgumentException(String argument, Frame frame, String column_name) { this(argument, (null == frame._key ? "null" : frame._key.toString()), column_name); } public H2OColumnNotFoundArgumentException(String argument, String frame_name, String column_name) { super("Column: " + column_name + " not found in frame: " + frame_name + " from argument: " + argument + ": " + argument.toString(), "Column: " + column_name + " not found in frame: " + frame_name + " from argument: " + argument + ": " + argument.toString()); this.values = new IcedHashMapGeneric.IcedHashMapStringObject(); this.values.put("argument", argument); this.values.put("frame_name", frame_name); this.values.put("column_name", column_name); } public H2OColumnNotFoundArgumentException(Frame frame, String column_name) { this((null == frame._key ? "null" : frame._key.toString()), column_name); } public H2OColumnNotFoundArgumentException(String frame_name, String column_name) { super("Column: " + column_name + " not found in frame: " + frame_name + ".", "Column: " + column_name + " not found in frame: " + frame_name + "."); this.values = new IcedHashMapGeneric.IcedHashMapStringObject(); this.values.put("frame_name", frame_name); this.values.put("column_name", column_name); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OConcurrentModificationException.java
package water.exceptions; /** * H2OConcurrentModificationException signals that object was modified while being used in another * operation. * Example use case is deleting a Vec while a checksum is being calculated on it. */ public class H2OConcurrentModificationException extends H2OAbstractRuntimeException { public H2OConcurrentModificationException(String message) { super(message, message); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OFailException.java
package water.exceptions; import water.H2OError; import water.util.HttpResponseStatus; /** * RuntimeException which causes H2O to shut down. This should only be used for cases in * which the code is bad, for example because a case isn't covered which must be for the * product to function correctly, and which therefore should be caught quickly in the * development process. */ public class H2OFailException extends H2OAbstractRuntimeException { protected int HTTP_RESPONSE_CODE() { return HttpResponseStatus.INTERNAL_SERVER_ERROR.getCode(); } public H2OFailException(String message) { super(message, message); this.timestamp = System.currentTimeMillis(); } public H2OFailException(String msg, Throwable cause) { this(msg); this.initCause(cause); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OFileAccessDeniedException.java
package water.exceptions; /** * Exception thrown when a file matches file_deny_glob */ public class H2OFileAccessDeniedException extends H2OAbstractRuntimeException { public H2OFileAccessDeniedException(String message, String dev_message) { super(message, dev_message); } public H2OFileAccessDeniedException(String message) { super(message, message); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OGridException.java
package water.exceptions; import water.H2OError; public class H2OGridException extends H2OAbstractRuntimeException { private final Throwable _rootCause; private final int _httpResponse; public H2OGridException(String msg) { this(msg, null); } public H2OGridException(String msg, Throwable rootException) { this(msg, rootException, 0); } public H2OGridException(String msg, Throwable rootException, int httpResponse) { super(msg, msg); _rootCause = rootException; _httpResponse = httpResponse > 0 ? httpResponse : super.HTTP_RESPONSE_CODE(); } @Override protected int HTTP_RESPONSE_CODE() { return _httpResponse; } @Override public H2OError toH2OError(String error_url) { H2OError err; String rootMessage = _rootCause== null ? null : _rootCause.getMessage().trim(); if (_rootCause instanceof H2OAbstractRuntimeException) { err = ((H2OAbstractRuntimeException) _rootCause).toH2OError(error_url); } else { err = new H2OError(timestamp, error_url, getMessage(), dev_message, HTTP_RESPONSE_CODE(), values, _rootCause); } StringBuilder msg = new StringBuilder(getMessage().trim()); if (msg.charAt(msg.length()-1) != '.') msg.append('.'); if (rootMessage != null) { msg.append(' '); msg.append("Root cause: "); msg.append(rootMessage); err._exception_msg = msg.toString(); } return err; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OIllegalArgumentException.java
package water.exceptions; import water.util.HttpResponseStatus; import water.util.IcedHashMap; import water.util.IcedHashMapGeneric; public class H2OIllegalArgumentException extends H2OAbstractRuntimeException { protected int HTTP_RESPONSE_CODE() { return HttpResponseStatus.PRECONDITION_FAILED.getCode(); } public H2OIllegalArgumentException(String argument, String function, Object value) { super("Illegal argument: " + argument + " of function: " + function + ": " + (value == null ? "null":value.toString()), "Illegal argument: " + argument + " of function: " + function + ": " + (value == null ? "null":value.toString()) + " of class: " + (value == null ? "null":value.getClass())); this.values = new IcedHashMapGeneric.IcedHashMapStringObject(); this.values.put("function", function); this.values.put("argument", argument); if (value!=null) { try { this.values.put("value", value); } catch (Exception ignored) {} } } /** Raw-message constructor for use by subclasses. */ public H2OIllegalArgumentException(String message, String dev_message, IcedHashMapGeneric.IcedHashMapStringObject values) { super(message, dev_message, values); } /** Raw-message constructor for use by subclasses. */ public H2OIllegalArgumentException(String message, String dev_message) { super(message, dev_message); } /** Raw-message constructor for use by subclasses. */ public H2OIllegalArgumentException(String message) { super(message, message); } public static H2OIllegalArgumentException wrongKeyType(String fieldName, String keyName, String expectedType, Class actualType) { H2OIllegalArgumentException e = new H2OIllegalArgumentException( expectedType + " argument: " + fieldName + " with value: " + keyName + " points to a non-" + expectedType + " object: " + actualType.getSimpleName(), expectedType + " argument: " + fieldName + " with value: " + keyName + " points to a non-" + expectedType + " object: " + actualType.getName()); e.values = new IcedHashMapGeneric.IcedHashMapStringObject(); e.values.put("argument", fieldName); e.values.put("value", keyName); e.values.put("expected_type", expectedType); e.values.put("actual_type", actualType); return e; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OIllegalValueException.java
package water.exceptions; import water.util.IcedHashMap; import water.util.IcedHashMapGeneric; /** * Exception indicating that we found an illegal value which was not passed in as an argument. */ public class H2OIllegalValueException extends H2OAbstractRuntimeException { public H2OIllegalValueException(String field, String object, Object value) { super("Illegal value for field: " + field + " of object: " + object + ": " + value.toString(), "Illegal value for field: " + field + " of object: " + object + ": " + value.toString() + " of class: " + value.getClass()); this.values = new IcedHashMapGeneric.IcedHashMapStringObject(); this.values.put("field", field); this.values.put("object", object); this.values.put("value", value); } public H2OIllegalValueException(String field, Object value) { super("Illegal value for field: " + field + ": " + value.toString(), "Illegal value for field: " + field + ": " + value.toString() + " of class: " + value.getClass()); this.values = new IcedHashMapGeneric.IcedHashMapStringObject(); this.values.put("field", field); this.values.put("value", value); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OKeyNotFoundArgumentException.java
package water.exceptions; import water.Key; import water.util.IcedHashMap; import water.util.IcedHashMapGeneric; /** * Exception signalling that a Key was not found. * <p> * If the Key name came from an argument, especially from an API parameter, use * {@code H2OKeyNotFoundArgumentException(String argument, String name)} or {@code H2OKeyNotFoundArgumentException(String argument, Key key)}, * which let you specify the argument name. If not, use {@code H2OKeyNotFoundArgumentException(String argument, String name)} or * {@code H2OKeyNotFoundArgumentException(String argument, Key key)}. */ public class H2OKeyNotFoundArgumentException extends H2ONotFoundArgumentException { public H2OKeyNotFoundArgumentException(String argument, String function, String name) { super("Object '" + name.toString() + "' not found in function: " + function + " for argument: " + argument, "Object '" + name.toString() + "' not found in function: " + function + " for argument: " + argument); this.values = new IcedHashMapGeneric.IcedHashMapStringObject(); this.values.put("function", function); this.values.put("argument", argument); this.values.put("name", name); } public H2OKeyNotFoundArgumentException(String argument, String name) { super("Object '" + name.toString() + "' not found for argument: " + argument, "Object '" + name.toString() + "' not found for argument: " + argument); this.values = new IcedHashMapGeneric.IcedHashMapStringObject(); this.values.put("argument", argument); this.values.put("name", name); } public H2OKeyNotFoundArgumentException(String argument, Key key) { this(argument, null == key ? "null" : key.toString()); } public H2OKeyNotFoundArgumentException(String name) { super("Object not found: " + name.toString(), "Object not found: " + name.toString()); this.values = new IcedHashMapGeneric.IcedHashMapStringObject(); this.values.put("name", name); } public H2OKeyNotFoundArgumentException(Key key) { this(null == key ? "null" : key.toString()); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OKeyWrongTypeArgumentException.java
package water.exceptions; import water.util.HttpResponseStatus; import water.Keyed; import water.util.IcedHashMap; import water.util.IcedHashMapGeneric; public class H2OKeyWrongTypeArgumentException extends H2OIllegalArgumentException { protected int HTTP_RESPONSE_CODE() { return HttpResponseStatus.NOT_FOUND.getCode(); } public H2OKeyWrongTypeArgumentException(String argument, Object value, Class<? extends Keyed> expected, Class actual) { super("Expected a " + expected.getSimpleName() + " for key argument: " + argument + " with value: " + value + ". Found a: " + actual.getSimpleName(), "Expected a " + expected.getCanonicalName() + " for key argument: " + argument + " with value: " + value + ". Found a: " + actual.getCanonicalName()); this.values = new IcedHashMapGeneric.IcedHashMapStringObject(); this.values.put("argument", argument); this.values.put("value", value); this.values.put("expected_type", expected.getCanonicalName()); this.values.put("actual_type", actual.getCanonicalName()); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OKeysNotFoundArgumentException.java
package water.exceptions; import water.util.IcedHashMap; import water.util.IcedHashMapGeneric; public class H2OKeysNotFoundArgumentException extends H2ONotFoundArgumentException { public H2OKeysNotFoundArgumentException(String argument, String[] names) { super("Objects not found: " + argument + ": " + names.toString(), "Objects not found: " + argument + ": " + names.toString()); this.values = new IcedHashMapGeneric.IcedHashMapStringObject(); this.values.put("argument", argument); this.values.put("names", names); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OModelBuilderIllegalArgumentException.java
package water.exceptions; import hex.Model; import hex.ModelBuilder; import water.H2OModelBuilderError; import water.util.IcedHashMapGeneric; public class H2OModelBuilderIllegalArgumentException extends H2OIllegalArgumentException { /** Raw-message constructor for use by the factory method. */ private H2OModelBuilderIllegalArgumentException(String message, String dev_message) { super(message, dev_message); } public static H2OModelBuilderIllegalArgumentException makeFromBuilder(ModelBuilder builder) { Model.Parameters parameters = builder._parms; String algo = builder._parms.algoName(); String msg = "Illegal argument(s) for " + algo + " model: " + builder.dest() + ". Details: " + builder.validationErrors(); H2OModelBuilderIllegalArgumentException exception = new H2OModelBuilderIllegalArgumentException(msg, msg); exception.values = new IcedHashMapGeneric.IcedHashMapStringObject(); exception.values.put("algo", algo); exception.values.put("parameters", parameters); exception.values.put("error_count", builder.error_count()); exception.values.put("messages", builder._messages); return exception; } public H2OModelBuilderError toH2OError(String error_url) { return new H2OModelBuilderError(timestamp, error_url, getMessage(), dev_message, HTTP_RESPONSE_CODE(), values, this); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2ONotFoundArgumentException.java
package water.exceptions; import water.util.HttpResponseStatus; public class H2ONotFoundArgumentException extends H2OIllegalArgumentException { final protected int HTTP_RESPONSE_CODE() { return HttpResponseStatus.NOT_FOUND.getCode(); } public H2ONotFoundArgumentException(String msg, String dev_msg) { super(msg, dev_msg); } public H2ONotFoundArgumentException(String msg) { super(msg, msg); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/H2OUnsupportedDataFileException.java
package water.exceptions; import water.util.IcedHashMapGeneric; /** * Exception thrown by a parser when a file format is recognized but a certain feature used * in the particular data file is not supported (eg. nested structures). */ public class H2OUnsupportedDataFileException extends H2OAbstractRuntimeException { public H2OUnsupportedDataFileException(String message, String dev_message, IcedHashMapGeneric.IcedHashMapStringObject values) { super(message, dev_message, values); } public H2OUnsupportedDataFileException(String message, String dev_message) { super(message, dev_message, new IcedHashMapGeneric.IcedHashMapStringObject()); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/exceptions/JCodeSB.java
package water.exceptions; import water.util.IcedBitSet; /** * Simple interface to share definition of SB and SBPrintStream. * * Designed for Java code generation. */ public interface JCodeSB<T extends JCodeSB<T>> { // Append primitves T ps(String s); T p(String s); //T p(float s); - intentionally not included to avoid issues with confusing p(float) and pj(float) - we don't want to output floats that look like doubles in Java code T p(double s); T p(char s); T p(int s); T p(long s); T p(boolean s); T p(JCodeSB s); T pobj(Object s); /** Increase indentation counter */ T i(int d); /** Increase indentation counter */ T i(); /** Indent and append. */ T ip(String s); /** Append empty string. */ T s(); // Java specific append of double T pj(double s); // Java specific append of float T pj(float s); /* Append Java string - escape all " and \ */ T pj(String s); /** Append reference to object's field * * @param objectName name of object * @param fieldName field name to reference * @return */ T pj(String objectName, String fieldName); T p(IcedBitSet ibs); /** Increase indentation counter */ T ii(int i); /** Decrease indentation counter */ T di(int i); // Copy indent from given string buffer T ci(JCodeSB sb); T nl(); // Convert a String[] into a valid Java String initializer T toJavaStringInit(String[] ss); T toJavaStringInit(float[] ss); T toJavaStringInit(double[] ss); T toJavaStringInit(double[][] ss); T toJavaStringInit(double[][][] ss); T toJSArray(float[] nums); T toJSArray(String[] ss); int getIndent(); String getContent(); }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/AppendableVec.java
package water.fvec; import water.*; import java.util.Arrays; /** * A NEW single distributed vector column. * * The NEW vector has no data, and takes no space. It supports distributed * parallel writes to it, via calls to append2. Such writes happen in parallel * and all writes are ordered. Writes *will* be local to the node doing them, * specifically to allow control over locality. By default, writes will go * local-homed chunks with no compression; there is a final 'close' to the NEW * vector which may do compression; the final 'close' will return some other * Vec type. NEW Vectors do NOT support reads! */ public class AppendableVec extends Vec { // Temporary ESPC, for uses which do not know the number of Chunks up front. public long _tmp_espc[]; // Allow Chunks to have their final Chunk index (set at closing) offset by // this much. Used by the Parser to fold together multi-file AppendableVecs. public final int _chunkOff; public AppendableVec( Key<Vec> key, byte type ) { this(key, new long[4], type, 0); } public AppendableVec( Key<Vec> key, long[] tmp_espc, byte type, int chunkOff) { super( key, -1/*no rowLayout yet*/, null, type ); _tmp_espc = tmp_espc; _chunkOff = chunkOff; } // A NewVector chunk was "closed" - completed. Add it's info to the roll-up. // This call is made in parallel across all node-local created chunks, but is // not called distributed. synchronized void closeChunk( int cidx, int len ) { // The Parser will pre-allocate the _tmp_espc large enough (the Parser // knows how many final Chunks there will be up front). Other users are // encouraged to set a "large enough" espc - and a shared one at that - to // avoid these copies. // Set the length into the temp ESPC at the Chunk index (accounting for _chunkOff) cidx -= _chunkOff; while( cidx >= _tmp_espc.length ) // should not happen if espcs are preallocated and shared! _tmp_espc = Arrays.copyOf(_tmp_espc, _tmp_espc.length<<1); _tmp_espc[cidx] = len; } public static Vec[] closeAll(AppendableVec [] avs) { Futures fs = new Futures(); Vec [] res = closeAll(avs,fs); fs.blockForPending(); return res; } public static Vec[] closeAll(AppendableVec [] avs, Futures fs) { Vec [] res = new Vec[avs.length]; final int rowLayout = avs[0].compute_rowLayout(); for(int i = 0; i < avs.length; ++i) res[i] = avs[i].close(rowLayout,fs); return res; } // Class 'reduce' call on new vectors; to combine the roll-up info. // Called single-threaded from the M/R framework. public void reduce( AppendableVec nv ) { if( this == nv ) return; // Trivially done if( _tmp_espc == nv._tmp_espc ) return; // Combine arrays of elements-per-chunk long e1[] = nv._tmp_espc; // Shorter array of longs? if (e1.length > _tmp_espc.length) { // Keep longer array e1 = _tmp_espc; // Keep the shorter one in e1 _tmp_espc = nv._tmp_espc; // Keep longer in the object } for( int i=0; i<e1.length; i++ ) // Copy non-zero elements over if( _tmp_espc[i]==0 && e1[i] != 0 ) // Read-filter (old code unconditionally did a R-M-W cycle) _tmp_espc[i] = e1[i]; // Only write if needed } public Vec layout_and_close(Futures fs) { return close(compute_rowLayout(),fs); } public int compute_rowLayout() { int nchunk = _tmp_espc.length; while( nchunk > 1 && _tmp_espc[nchunk-1] == 0 ) nchunk--; // Compute elems-per-chunk. // Roll-up elem counts, so espc[i] is the starting element# of chunk i. long espc[] = new long[nchunk+1]; // Shorter array long x=0; // Total row count so far for( int i=0; i<nchunk; i++ ) { espc[i] = x; // Start elem# for chunk i x += _tmp_espc[i]; // Raise total elem count } espc[nchunk]=x; // Total element count in last return ESPC.rowLayout(_key,espc); } // "Close" out a NEW vector - rewrite it to a plain Vec that supports random // reads, plus computes rows-per-chunk, min/max/mean, etc. public Vec close(int rowLayout, Futures fs) { // Compute #chunks int nchunk = _tmp_espc.length; DKV.remove(chunkKey(nchunk),fs); // remove potential trailing key while( nchunk > 1 && _tmp_espc[nchunk-1] == 0 ) { nchunk--; DKV.remove(chunkKey(nchunk),fs); // remove potential trailing key } // Replacement plain Vec for AppendableVec. Vec vec = new Vec(_key, rowLayout, domain(), _type); DKV.put(_key,vec,fs); // Inject the header into the K/V store return vec; } // Default read/write behavior for AppendableVecs @Override protected boolean readable() { return false; } @Override protected boolean writable() { return true ; } @Override public NewChunk chunkForChunkIdx(int cidx) { return new NewChunk(this,cidx); } // None of these are supposed to be called while building the new vector @Override public Value chunkIdx( int cidx ) { throw H2O.fail(); } @Override public long length() { throw H2O.fail(); } @Override public int nChunks() { throw H2O.fail(); } @Override public int nonEmptyChunks() { throw H2O.fail(); } @Override public int elem2ChunkIdx( long i ) { throw H2O.fail(); } @Override protected long chunk2StartElem( int cidx ) { throw H2O.fail(); } @Override public long byteSize() { return 0; } @Override public String toString() { return "[AppendableVec, unknown size]"; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/ByteVec.java
package water.fvec; import water.Job; import water.Key; import water.Value; import java.io.IOException; import java.io.InputStream; /** * A vector of plain Bytes. */ public class ByteVec extends Vec { public ByteVec( Key key, int rowLayout ) { super(key, rowLayout); } @Override public C1NChunk chunkForChunkIdx(int cidx) { return (C1NChunk)super.chunkForChunkIdx(cidx); } /** Return column missing-element-count - ByteVecs do not allow any "missing elements" */ @Override public long naCnt() { return 0; } /** Is all integers? Yes, it's all bytes */ @Override public boolean isInt(){return true; } /** Get an unspecified amount of initial bytes; typically a whole C1NChunk of * length Vec.DFLT_CHUNK_SIZE but no guarantees. Useful for previewing the start * of large files. * @return array of initial bytes */ public byte[] getFirstBytes() { return getFirstChunkBytes(); } final byte[] getFirstChunkBytes() { return chunkForChunkIdx(0)._mem; } public InputStream openStream() { return openStream(null); } /** * Open a stream view over the underlying data */ public InputStream openStream(final Key job_key) { InputStream is = new InputStream() { final long[] sz = new long[1]; private int _cidx, _pidx, _sz; private C1NChunk _c0; @Override public int available() { if (_c0 == null || _sz >= _c0._len) { sz[0] += _c0 != null ? _c0._len : 0; if (_cidx >= nChunks()) return 0; _c0 = chunkForChunkIdx(_cidx++); _sz = C1NChunk._OFF; if (job_key != null) Job.update(_c0._len, job_key); } return _c0._len - _sz; } @Override public void close() { _cidx = nChunks(); _c0 = null; _sz = 0; } @Override public int read() throws IOException { return available() == 0 ? -1 : 0xFF & _c0._mem[_sz++]; } @Override public int read(byte[] b, int off, int len) { if (b == null) { // Back-channel read of cidx if (_cidx > _pidx) { // Remove prev chunk from memory Value v = Value.STORE_get(chunkKey(_pidx++)); if (v != null && v.isPersisted()) { v.freePOJO(); // Eagerly toss from memory v.freeMem(); } // Else not found, or not on disk somewhere } return _cidx; } int sz = available(); if (sz == 0) return -1; len = Math.min(len, sz); System.arraycopy(_c0._mem, _sz, b, off, len); _sz += len; return len; } }; try { is.available(); } catch (IOException e) { throw new RuntimeException(e); } return is; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C0DChunk.java
package water.fvec; import water.parser.BufferedString; import water.util.UnsafeUtils; import java.util.UUID; /** * The constant 'double' column (also used to represent constant NA chunks). */ public class C0DChunk extends Chunk { private static final int _OFF=8+4; private double _con; public C0DChunk(double con, int len) { _start = -1; set_len(len); _mem=new byte[_OFF]; _con = con; UnsafeUtils.set8d(_mem, 0, con); UnsafeUtils.set4(_mem,8,len); } @Override public boolean hasFloat() { return ! Double.isNaN(_con); /* Constant NA chunk is not Float! */ } @Override protected final long at8_impl( int i ) { if( Double.isNaN(_con) ) throw new IllegalArgumentException("at8_abs but value is missing"); return (long)_con; // Possible silent truncation } long at16h_impl(int idx) { throw wrongType(UUID.class, Object.class); } @Override protected final double atd_impl( int i ) {return _con;} @Override protected final boolean isNA_impl( int i ) { return Double.isNaN(_con); } @Override boolean set_impl(int idx, long l) { return l==_con; } @Override boolean set_impl(int i, double d) { return d==_con; } @Override boolean set_impl(int i, float f ) { return f==_con; } @Override boolean setNA_impl(int i) { return Double.isNaN(_con); } @Override double min() { return _con; } @Override double max() { return _con; } BufferedString atStr_impl(BufferedString bStr, int idx) { if(Double.isNaN(_con)) return null; // speciall all missing case return super.atStr_impl(bStr,idx); } // 3.3333333e33 // public int pformat_len0() { return 22; } // public String pformat0() { return "% 21.15e"; } // Custom serializers: the _mem field contains ALL the fields already. // Init _start to -1, so we know we have not filled in other fields. // Leave _vec & _chk2 null, leave _len unknown. @Override final public void initFromBytes() { _start = -1; _cidx = -1; _con = UnsafeUtils.get8d(_mem,0); set_len(UnsafeUtils.get4(_mem,8)); } @Override public boolean isSparseZero(){return _con == 0;} @Override public int sparseLenZero() {return _con ==0?0:_len;} @Override public int nextNZ(int rid) {return _con==0?_len:rid+1;} @Override public int nonzeros(int [] arr) { if (_con == 0) return 0; for (int i = 0; i < _len; ++i) arr[i] = i; return _len; } @Override public boolean isSparseNA(){return Double.isNaN(_con);} @Override public int sparseLenNA() {return Double.isNaN(_con)?0:_len;} @Override public int getSparseDoubles(double [] vals, int [] ids, double NA){ if(_con == 0) return 0; double con = Double.isNaN(_con)?NA:_con; for(int i = 0; i < _len; ++i) { vals[i] = con; ids[i] = i; } return _len; } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to){ if(_con == 0) v.addZeros(to-from); else if(Double.isNaN(_con)) v.addNAs(to-from); else for(int i = from; i < to; i++) v.addValue(_con); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int [] ids){ if(_con == 0) v.addZeros(ids.length); else if(Double.isNaN(_con)) v.addNAs(ids.length); else for(int i = 0; i < ids.length; i++) v.addValue(_con); return v; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C0LChunk.java
package water.fvec; import water.util.UnsafeUtils; /** * The constant 'long' column. */ public class C0LChunk extends Chunk { protected static final int _OFF=8+4; private long _con; public C0LChunk(long con, int len) { _mem=new byte[_OFF]; _start = -1; set_len(len); _con = con; UnsafeUtils.set8(_mem, 0, con); UnsafeUtils.set4(_mem,8,len); } @Override public boolean hasFloat() { return false; } @Override public boolean hasNA() { return false; } @Override protected final long at8_impl( int i ) { return _con; } @Override protected final double atd_impl( int i ) {return _con; } @Override protected final boolean isNA_impl( int i ) { return false; } @Override boolean set_impl(int idx, long l) { return l==_con; } @Override boolean set_impl(int i, double d) { return d==_con; } @Override boolean set_impl(int i, float f ) { return f==_con; } @Override boolean setNA_impl(int i) { return false; } @Override boolean set_impl (int idx, String str) { return false; } @Override double min() { return _con; } @Override double max() { return _con; } @Override public final void initFromBytes () { _start = -1; _cidx = -1; _con = UnsafeUtils.get8(_mem,0); set_len(UnsafeUtils.get4(_mem,8)); } @Override public boolean isSparseZero(){return _con == 0;} @Override public int sparseLenZero(){return _con == 0?0: _len;} @Override public int nextNZ(int rid){return _con == 0?_len:rid+1;} @Override public int nonzeros(int [] arr) { if (_con == 0) return 0; for (int i = 0; i < _len; ++i) arr[i] = i; return _len; } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to){ if(_con == 0) v.addZeros(to-from); else for(int i = from; i < to; i++) v.addValue(_con); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int [] ids){ if(_con == 0) v.addZeros(ids.length); else for(int i = 0; i < ids.length; i++) v.addValue(_con); return v; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C16Chunk.java
package water.fvec; import water.util.UnsafeUtils; public class C16Chunk extends Chunk { static final long _LO_NA = Long.MIN_VALUE; static final long _HI_NA = 0; C16Chunk( byte[] bs ) { _mem=bs; _start = -1; set_len(_mem.length>>4); } @Override protected final long at8_impl( int i ) { throw new IllegalArgumentException("at8_abs but 16-byte UUID"); } @Override protected final double atd_impl( int i ) { throw new IllegalArgumentException("atd but 16-byte UUID"); } @Override protected final boolean isNA_impl( int i ) { return isNA(loAt(i), hiAt(i)); } public static boolean isNA(long lo, long hi) { return lo ==_LO_NA && hi ==_HI_NA; } private long loAt(int idx) { return UnsafeUtils.get8(_mem, idx*16); } private long hiAt(int idx) { return UnsafeUtils.get8(_mem, idx*16+8); } @Override protected long at16l_impl(int idx) { long lo = loAt(idx); if (lo == _LO_NA && hiAt(idx) == _HI_NA) { throw new IllegalArgumentException("at16l but value is missing at " + idx); } return lo; } @Override protected long at16h_impl(int idx) { long hi = hiAt(idx); if (hi == _HI_NA && loAt(idx) == _LO_NA) { throw new IllegalArgumentException("at16h but value is missing at " + idx); } return hi; } @Override boolean set_impl(int i, long lo, long hi) { if (isNA(lo, hi)) throw new IllegalArgumentException("Illegal uid value"); UnsafeUtils.set8(_mem, i*16, lo); UnsafeUtils.set8(_mem, i*16 + 8, hi); return true; } @Override boolean set_impl(int idx, long l) { return false; } @Override boolean set_impl(int i, double d) { return false; } @Override boolean set_impl(int i, float f ) { return false; } @Override boolean setNA_impl(int idx) { return set_impl(idx, _LO_NA, _HI_NA); } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to) { for(int i = from; i < to; i++) { if(isNA(i)) v.addNAs(1); else v.addValue(UnsafeUtils.get8(_mem, 16 * i), UnsafeUtils.get8(_mem, 16 * i + 8)); } return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int[] ids) { for(int i:ids) v.addValue(UnsafeUtils.get8(_mem,16*i),UnsafeUtils.get8(_mem,16*i+8)); return v; } @Override protected final void initFromBytes () { _start = -1; _cidx = -1; set_len(_mem.length>>4); assert _mem.length == _len * 16; } // @Override protected int pformat_len0() { return 36; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C1Chunk.java
package water.fvec; /** * The empty-compression function, if all elements fit directly on UNSIGNED bytes. * Cannot store 0xFF, the value is a marker for N/A. */ public class C1Chunk extends Chunk { static protected final int _OFF = 0; static protected final int _NA = 0xFF; C1Chunk(byte[] bs) { _mem=bs; _start = -1; set_len(_mem.length); } @Override protected final long at8_impl( int i ) { long res = 0xFF&_mem[i+_OFF]; if( res == _NA ) throw new IllegalArgumentException("at8_abs but value is missing"); return res; } @Override protected final double atd_impl( int i ) { long res = 0xFF&_mem[i+_OFF]; return (res == _NA)?Double.NaN:res; } @Override protected final boolean isNA_impl( int i ) { return (0xFF&_mem[i+_OFF]) == _NA; } @Override boolean set_impl(int i, long l) { if( !(0 <= l && l < 255) ) return false; _mem[i+_OFF] = (byte)l; return true; } @Override boolean set_impl(int i, double d) { return false; } @Override boolean set_impl(int i, float f ) { return false; } @Override boolean setNA_impl(int idx) { _mem[idx+_OFF] = (byte)_NA; return true; } @Override public void initFromBytes(){ _start = -1; _cidx = -1; set_len(_mem.length); } private final void processRow(int r, ChunkVisitor v){ int i = 0xFF&_mem[r+_OFF]; if(i == _NA) v.addNAs(1); else v.addValue(i); } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to) { for(int i = from; i < to; i++) processRow(i,v); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int[] ids) { for(int i:ids) processRow(i,v); return v; } public int [] getIntegers(int [] vals, int from, int to, int NA){ for(int i = from; i < to; i++) { int x = 0xFF&_mem[i]; vals[i-from] = (x == _NA)?NA:x; } return vals; } @Override public double [] getDoubles(double [] vals, int from, int to, double NA){ for(int i = from; i < to; i++) { int x = 0xFF&_mem[i]; vals[i-from] = (x == _NA)?NA:x; } return vals; } @Override public double [] getDoubles(double [] vals, int [] ids){ int k = 0; for(int i:ids) { int x = 0xFF&_mem[i]; vals[k++] = (x == _NA)?Double.NaN:x; } return vals; } @Override public boolean hasFloat() {return false;} }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C1NChunk.java
package water.fvec; /** * The empty-compression function, if all elements fit directly on UNSIGNED bytes. * [In particular, this is the compression style for data read in from files.] */ public class C1NChunk extends Chunk { protected static final int _OFF=0; public C1NChunk(byte[] bs) { _mem=bs; _start = -1; set_len(_mem.length); } @Override protected final long at8_impl( int i ) { return 0xFF&_mem[i]; } @Override protected final double atd_impl( int i ) { return 0xFF&_mem[i]; } @Override protected final boolean isNA_impl( int i ) { return false; } @Override boolean set_impl(int i, long l ) { return false; } @Override boolean set_impl(int i, double d) { return false; } @Override boolean set_impl(int i, float f ) { return false; } @Override boolean setNA_impl(int idx) { return false; } // Custom serializers: the _mem field contains ALL the fields already. // Init _start to -1, so we know we have not filled in other fields. // Leave _vec & _chk2 null, leave _len unknown. @Override protected final void initFromBytes () { _start = -1; _cidx = -1; set_len(_mem.length); } @Override public boolean hasFloat() {return false;} @Override public boolean hasNA() { return false; } @Override public double [] getDoubles(double [] vals, int [] ids) { int k = 0; for (int i : ids) vals[k++] = _mem[i] & 0xFF; return vals; } @Override public double [] getDoubles(double [] vals, int from, int to, double NA){ for(int i = from; i < to; ++i) vals[i-from] = _mem[i]&0xFF; return vals; } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to) { for(int i = from; i < to; i++) v.addValue(0xFF&_mem[i]); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int[] ids) { for(int i:ids) v.addValue(0xFF&_mem[i]); return v; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C1SChunk.java
package water.fvec; import water.util.PrettyPrint; /** * The scale/bias function, where data is in SIGNED bytes before scaling. */ public final class C1SChunk extends CSChunk { C1SChunk( byte[] bs, long bias, int scale) { super(bs,bias,scale,0); if(scale < 0) { // check precision double div = PrettyPrint.pow10(1, -scale); for (int i = 0; i < _len; ++i) { int x = 0xFF & _mem[_OFF + i]; if (x == C1Chunk._NA) continue; if ((getD(x, C1Chunk._NA, Double.NaN)) != (x+bias)/div){ setDecimal(); break; } } } } @Override protected final long at8_impl( int i ) { int x = 0xFF&_mem[_OFF+i]; if( x==C1Chunk._NA ) throw new IllegalArgumentException("at8_abs but value is missing"); return get8(x); } @Override protected final double atd_impl( int i ) { return getD(0xFF&_mem[_OFF+i],C1Chunk._NA); } @Override protected final boolean isNA_impl( int i ) { return (0xFF&_mem[i+_OFF]) == C1Chunk._NA; } @Override boolean setNA_impl(int idx) { _mem[idx+_OFF] = (byte)C1Chunk._NA; return true; } @Override protected boolean set_impl(int i, double x) { if(Double.isNaN(x)) return setNA_impl(i); int y = getScaledValue(x, C1Chunk._NA); byte b = (byte)y; if(getD(0xFF&b,C1Chunk._NA,Double.NaN) != x) return false; _mem[_OFF+i] = b; assert !isNA_impl(i); return true; } /** * Dense bulk interface, fetch values from the given range * @param vals * @param from * @param to */ @Override public double [] getDoubles(double [] vals, int from, int to, double NA){ for(int i = from; i < to; ++i) vals[i-from] = getD(0xFF&_mem[_OFF+i],C1Chunk._NA,NA); return vals; } /** * Dense bulk interface, fetch values from the given ids * @param vals * @param ids */ @Override public double [] getDoubles(double [] vals, int [] ids){ int j = 0; for(int i:ids) vals[j++] = getD(0xFF&_mem[_OFF+i],C1Chunk._NA); return vals; } private <T extends ChunkVisitor> void processRow(T v, int i, long bias, int exp){ long x = 0xFF & _mem[_OFF + i]; if(x == C1Chunk._NA) v.addNAs(1); else v.addValue(x + bias, exp); } @Override protected <T extends ChunkVisitor> T processRows2(T v, int from, int to, long bias, int exp) { for(int i = from; i < to; ++i) processRow(v,i,bias,exp); return v; } @Override protected <T extends ChunkVisitor> T processRows2(T v, int from, int to) { for(int i = from; i < to; ++i) v.addValue(getD(0xFF&_mem[_OFF+i],C1Chunk._NA)); return v; } @Override protected <T extends ChunkVisitor> T processRows2(T v, int [] ids, long bias, int exp) { for(int i:ids) processRow(v,i,bias,exp); return v; } @Override protected <T extends ChunkVisitor> T processRows2(T v, int [] ids) { for(int i:ids) v.addValue(getD(0xFF&_mem[_OFF+i],C1Chunk._NA)); return v; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C2Chunk.java
package water.fvec; import water.util.UnsafeUtils; /** * The empty-compression function, where data is in shorts. */ public class C2Chunk extends Chunk { static protected final int _NA = Short.MIN_VALUE; static protected final int _OFF=0; C2Chunk( byte[] bs ) { _mem=bs; _start = -1; set_len(_mem.length>>1); } @Override protected final long at8_impl( int i ) { int res = UnsafeUtils.get2(_mem,(i<<1)+_OFF); if( res == _NA ) throw new IllegalArgumentException("at8_abs but value is missing"); return res; } @Override protected final double atd_impl( int i ) { int res = UnsafeUtils.get2(_mem,(i<<1)+_OFF); return res == _NA?Double.NaN:res; } @Override protected final boolean isNA_impl( int i ) { return UnsafeUtils.get2(_mem,(i<<1)+_OFF) == _NA; } @Override boolean set_impl(int idx, long l) { if( !(Short.MIN_VALUE < l && l <= Short.MAX_VALUE) ) return false; UnsafeUtils.set2(_mem,(idx<<1)+_OFF,(short)l); return true; } @Override boolean set_impl(int idx, double d) { if( Double.isNaN(d) ) return setNA_impl(idx); long l = (long)d; return l == d && set_impl(idx, l); } @Override boolean set_impl(int i, float f ) { return set_impl(i,(double)f); } @Override boolean setNA_impl(int idx) { UnsafeUtils.set2(_mem,(idx<<1)+_OFF,(short)_NA); return true; } private final void processRow(int r, ChunkVisitor v){ int i = UnsafeUtils.get2(_mem,(r<<1)+_OFF); if(i == _NA) v.addNAs(1); else v.addValue(i); } @Override public double [] getDoubles(double [] vals, int from, int to, double NA){ for(int i = from; i < to; i++) { int x = UnsafeUtils.get2(_mem, 2*i); vals[i-from] = (x == _NA)?NA:x; } return vals; } @Override public double [] getDoubles(double [] vals, int [] ids){ int k = 0; for(int i:ids) { int x = UnsafeUtils.get2(_mem, 2*i); vals[k++] = (x == _NA)?Double.NaN:x; } return vals; } @Override public int [] getIntegers(int [] vals, int from, int to, int NA){ for(int i = from; i < to; i++) { int x = UnsafeUtils.get2(_mem, 2*i); vals[i-from] = (x == _NA)?NA:x; } return vals; } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to) { for(int i = from; i < to; i++) processRow(i,v); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int[] ids) { for(int i:ids) processRow(i,v); return v; } @Override public final void initFromBytes () { _start = -1; _cidx = -1; set_len(_mem.length>>1); assert _mem.length == _len <<1; } @Override public boolean hasFloat() {return false;} }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C2SChunk.java
package water.fvec; import water.util.PrettyPrint; import water.util.UnsafeUtils; /** * The scale/bias function, where data is in SIGNED bytes before scaling. */ public class C2SChunk extends CSChunk { C2SChunk( byte[] bs, long bias, int scale ) { super(bs,bias,scale,1); if(scale < 0) { // check precision double div = PrettyPrint.pow10(1, -scale); for (int i = 0; i < _len; ++i) { int x = getMantissa(i); if (x == C2Chunk._NA) continue; if ((getD(x, C2Chunk._NA, Double.NaN)) != (x+bias)/div){ setDecimal(); break; } } } } @Override protected final long at8_impl( int i ) { int x = getMantissa(i); if( x==C2Chunk._NA ) throw new IllegalArgumentException("at8_abs but value is missing"); return get8(x); } private int getMantissa(int i){return UnsafeUtils.get2(_mem,_OFF+2*i);} private void setMantissa(int i, short s){ UnsafeUtils.set2(_mem,(i*2)+_OFF,s); } @Override protected final double atd_impl( int i ) {return getD(getMantissa(i),C2Chunk._NA);} @Override protected final boolean isNA_impl( int i ) { return getMantissa(i) == C2Chunk._NA; } @Override boolean set_impl(int i, double x) { if(Double.isNaN(x)) return setNA_impl(i); int y = getScaledValue(x, C2Chunk._NA); short s = (short)y; if(getD(s,C2Chunk._NA, Double.NaN) != x) return false; setMantissa(i,s); assert !isNA_impl(i); return true; } @Override boolean setNA_impl(int idx) {setMantissa(idx,(short)C2Chunk._NA); return true; } @Override public double [] getDoubles(double [] vals, int from, int to, double NA){ for(int i = from; i < to; i++) vals[i-from] = getD(getMantissa(i),C2Chunk._NA,NA); return vals; } @Override public double [] getDoubles(double [] vals, int [] ids){ int k = 0; for(int i:ids) vals[k++] = getD(getMantissa(i),C2Chunk._NA); return vals; } private <T extends ChunkVisitor> void processRow(T v, int i, long bias, int exp){ long x = getMantissa(i); if(x == C2Chunk._NA) v.addNAs(1); else v.addValue(x + bias, exp); } @Override protected <T extends ChunkVisitor> T processRows2(T v, int from, int to, long bias, int exp) { for(int i = from; i < to; ++i) processRow(v,i,bias,exp); return v; } @Override protected <T extends ChunkVisitor> T processRows2(T v, int from, int to) { for(int i = from; i < to; ++i) v.addValue(getD(getMantissa(i),C2Chunk._NA)); return v; } @Override protected <T extends ChunkVisitor> T processRows2(T v, int [] ids, long bias, int exp) { for(int i:ids) processRow(v,i,bias,exp); return v; } @Override protected <T extends ChunkVisitor> T processRows2(T v, int [] ids) { for(int i:ids) v.addValue(getD(getMantissa(i),C2Chunk._NA)); return v; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C4Chunk.java
package water.fvec; import water.util.UnsafeUtils; /** * The empty-compression function, where data is in 'int's. */ public class C4Chunk extends Chunk { static protected final int _NA = Integer.MIN_VALUE; C4Chunk( byte[] bs ) { _mem=bs; _start = -1; set_len(_mem.length>>2); } @Override protected final long at8_impl( int i ) { long res = UnsafeUtils.get4(_mem,i<<2); if( res == _NA ) throw new IllegalArgumentException("at8_abs but value is missing"); return res; } @Override protected final double atd_impl( int i ) { long res = UnsafeUtils.get4(_mem, i << 2); return res == _NA?Double.NaN:res; } @Override protected final boolean isNA_impl( int i ) { return UnsafeUtils.get4(_mem,i<<2) == _NA; } @Override boolean set_impl(int idx, long l) { if( !(Integer.MIN_VALUE < l && l <= Integer.MAX_VALUE) ) return false; UnsafeUtils.set4(_mem,idx<<2,(int)l); return true; } @Override boolean set_impl(int i, double d) { return false; } @Override boolean set_impl(int i, float f ) { return false; } @Override boolean setNA_impl(int idx) { UnsafeUtils.set4(_mem,(idx<<2),(int)_NA); return true; } private final void processRow(int r, ChunkVisitor v){ int i = UnsafeUtils.get4(_mem,(r<<2)); if(i == _NA) v.addNAs(1); else v.addValue(i); } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to) { for(int i = from; i < to; i++) processRow(i,v); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int[] ids) { for(int i:ids) processRow(i,v); return v; } @Override public final void initFromBytes () { _start = -1; _cidx = -1; set_len(_mem.length>>2); assert _mem.length == _len <<2; } @Override public boolean hasFloat() {return false;} @Override public double [] getDoubles(double [] vals, int from, int to, double NA){ for(int i = from; i < to; i++) { int x = UnsafeUtils.get4(_mem, 4*i); vals[i-from] = (x == _NA)?NA:x; } return vals; } @Override public double [] getDoubles(double [] vals, int [] ids){ int k = 0; for(int i:ids) { int x = UnsafeUtils.get4(_mem, 4*i); vals[k++] = (x == _NA)?Double.NaN:x; } return vals; } @Override public int [] getIntegers(int [] vals, int from, int to, int NA){ for(int i = from; i < to; i++) { int x = UnsafeUtils.get4(_mem, 4*i); vals[i-from] = (x == _NA)?NA:x; } return vals; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C4FChunk.java
package water.fvec; import water.util.UnsafeUtils; /** * The empty-compression function, where data is in 'float's. */ public class C4FChunk extends Chunk { public C4FChunk( byte[] bs ) { _mem=bs; _start = -1; set_len(_mem.length>>2); } @Override protected final long at8_impl( int i ) { float res = UnsafeUtils.get4f(_mem, i << 2); if( Float.isNaN(res) ) throw new IllegalArgumentException("at8_abs but value is missing"); return (long)res; } @Override protected final double atd_impl( int i ) { float res = UnsafeUtils.get4f(_mem,i<<2); return Float.isNaN(res)?Double.NaN:res; } @Override protected final boolean isNA_impl( int i ) { return Float.isNaN(UnsafeUtils.get4f(_mem,i<<2)); } @Override boolean set_impl(int idx, long l) { return false; } @Override boolean set_impl(int i, double d) { return false; } @Override boolean set_impl(int i, float f ) { UnsafeUtils.set4f(_mem,i<<2,f); return true; } @Override boolean setNA_impl(int idx) { UnsafeUtils.set4f(_mem,(idx<<2),Float.NaN); return true; } @Override public NewChunk extractRows(NewChunk nc, int from, int to){ for(int i = from; i < to; i++) nc.addNum(UnsafeUtils.get4f(_mem,4*i)); return nc; } @Override public NewChunk extractRows(NewChunk nc, int... rows){ for(int i:rows) nc.addNum(UnsafeUtils.get4f(_mem,4*i)); return nc; } private final void processRow(int r, ChunkVisitor v){ float f = UnsafeUtils.get4f(_mem,(r<<2)); if(Float.isNaN(f)) v.addNAs(1); else v.addValue((double)f); } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to) { for(int i = from; i < to; i++) processRow(i,v); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int[] ids) { for(int i:ids) processRow(i,v); return v; } // 3.3333333e33 // public int pformat_len0() { return 14; } // public String pformat0() { return "% 13.7e"; } @Override public final void initFromBytes () { _start = -1; _cidx = -1; set_len(_mem.length>>2); assert _mem.length == _len <<2; } @Override public boolean hasFloat() {return true;} }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C4SChunk.java
package water.fvec; import water.util.PrettyPrint; import water.util.UnsafeUtils; /** * The scale/bias function, where data is in SIGNED bytes before scaling. */ public class C4SChunk extends CSChunk { C4SChunk( byte[] bs, long bias, int scale ) { super(bs,bias,scale,2); if(scale < 0) { // check precision double div = PrettyPrint.pow10(1, -scale); for (int i = 0; i < _len; ++i) { int x = getMantissa(i); if (x == C4Chunk._NA) continue; if ((getD(x, C1Chunk._NA, Double.NaN)) != (x+bias)/div){ setDecimal(); break; } } } } private int getMantissa(int i){return UnsafeUtils.get4(_mem,_OFF+4*i);} private void setMantissa(int i, int j){ UnsafeUtils.set4(_mem,(i*4)+_OFF,j); } @Override protected final double atd_impl( int i ) {return getD(getMantissa(i),C4Chunk._NA);} @Override protected final boolean isNA_impl( int i ) { return getMantissa(i) == C4Chunk._NA; } @Override boolean set_impl(int i, double x) { if(Double.isNaN(x)) return setNA_impl(i); int y = getScaledValue(x, C4Chunk._NA); if(getD(y,C4Chunk._NA, Double.NaN) != x) return false; setMantissa(i,y); assert !isNA_impl(i); return true; } @Override boolean setNA_impl(int idx) {setMantissa(idx,C4Chunk._NA); return true; } @Override public double [] getDoubles(double [] vals, int from, int to, double NA){ for(int i = from; i < to; i++) vals[i-from] = getD(getMantissa(i),C4Chunk._NA,NA); return vals; } @Override public double [] getDoubles(double [] vals, int [] ids){ int k = 0; for(int i:ids) vals[k++] = getD(getMantissa(i),C4Chunk._NA); return vals; } private <T extends ChunkVisitor> void processRow(T v, int i, long bias, int exp){ long x = getMantissa(i); if(x == C4Chunk._NA) v.addNAs(1); else v.addValue(x + bias, exp); } @Override protected <T extends ChunkVisitor> T processRows2(T v, int from, int to, long bias, int exp) { for(int i = from; i < to; ++i) processRow(v,i,bias,exp); return v; } @Override protected <T extends ChunkVisitor> T processRows2(T v, int from, int to) { for(int i = from; i < to; ++i) v.addValue(getD(getMantissa(i),C4Chunk._NA)); return v; } @Override protected <T extends ChunkVisitor> T processRows2(T v, int [] ids, long bias, int exp) { for(int i:ids) processRow(v,i,bias,exp); return v; } @Override protected <T extends ChunkVisitor> T processRows2(T v, int [] ids) { for(int i:ids) v.addValue(getD(getMantissa(i),C4Chunk._NA)); return v; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C4VolatileChunk.java
package water.fvec; import water.*; import water.util.UnsafeUtils; /** * The empty-compression function, where data is in 'int's. * Can only be used locally (intentionally does not serialize). * Intended for temporary data which gets modified frequently. * Exposes data directly as int[] */ public class C4VolatileChunk extends Chunk { static protected final long _NA = Integer.MIN_VALUE; transient private int [] _is; C4VolatileChunk(int[] is ) { _is = is; _mem = new byte[0]; _start = -1; _len = is.length; } public boolean isVolatile() {return true;} public int[] getValues(){return _is;} @Override protected final long at8_impl( int i ) { long res = _is[i]; if( res == _NA ) throw new IllegalArgumentException("at8_abs but value is missing"); return res; } @Override protected final double atd_impl( int i ) { long res = _is[i]; return res == _NA?Double.NaN:res; } @Override protected final boolean isNA_impl( int i ) { return _is[i] == _NA; } @Override boolean set_impl(int idx, long l) { if( !(Integer.MIN_VALUE < l && l <= Integer.MAX_VALUE) ) return false; _is[idx] = (int)l; return true; } @Override boolean set_impl(int i, double d) {return false; } @Override boolean set_impl(int i, float f ) {return false; } @Override boolean setNA_impl(int idx) { _is[idx] = (int)_NA; return true; } private final void processRow(int r, ChunkVisitor v){ int i = UnsafeUtils.get4(_mem,(r<<2)); if(i == _NA) v.addNAs(1); else v.addValue(i); } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to) { for(int i = from; i < to; i++) processRow(i,v); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int[] ids) { for(int i:ids) processRow(i,v); return v; } @Override public final void initFromBytes () { _len = _mem.length >> 2; _is = MemoryManager.malloc4(_len); for(int i = 0; i < _is.length; ++i) _is[i] = UnsafeUtils.get4(_mem,4*i); _mem = null; } @Override public Futures close( int cidx, Futures fs ) { if(chk2() != null) return chk2().close(cidx,fs); Value v = new Value(_vec.chunkKey(cidx),this,_len*4,Value.ICE); DKV.put(v._key,v,fs); return fs; } @Override public byte [] asBytes() { byte [] res = MemoryManager.malloc1(_len*4); for(int i = 0; i < _len; ++i) UnsafeUtils.set4(res,4*i,_is[i]); return res; } @Override public boolean hasFloat() {return false;} // public Futures close(int cidx, Futures fs ) { // always assume got modified // Value v = new Value(_vec.chunkKey(_cidx), this,this._len*4,Value.ICE); // DKV.put(v._key,v,fs); // Write updated chunk back into K/V // return fs; // } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C8Chunk.java
package water.fvec; import water.util.UnsafeUtils; /** * The empty-compression function, where data is in 'long's. */ public class C8Chunk extends Chunk { protected static final long _NA = Long.MIN_VALUE; C8Chunk( byte[] bs ) { _mem=bs; _start = -1; set_len(_mem.length>>3); } @Override protected final long at8_impl( int i ) { long res = UnsafeUtils.get8(_mem,i<<3); if( res == _NA ) throw new IllegalArgumentException("at8_abs but value is missing"); return res; } @Override protected final double atd_impl( int i ) { long res = UnsafeUtils.get8(_mem,i<<3); return res == _NA?Double.NaN:res; } @Override protected final boolean isNA_impl( int i ) { return UnsafeUtils.get8(_mem, i << 3)==_NA; } @Override boolean set_impl(int idx, long l) { return false; } @Override boolean set_impl(int i, double d) { return false; } @Override boolean set_impl(int i, float f ) { return false; } @Override boolean setNA_impl(int idx) { UnsafeUtils.set8(_mem,(idx<<3),_NA); return true; } @Override public final void initFromBytes () { _start = -1; _cidx = -1; set_len(_mem.length>>3); assert _mem.length == _len <<3; } @Override public boolean hasFloat() {return false;} private final void processRow(int r, ChunkVisitor v){ long l = UnsafeUtils.get8(_mem,(r<<3)); if(l == _NA) v.addNAs(1); else v.addValue(l); } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to) { for(int i = from; i < to; i++) processRow(i,v); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int[] ids) { for(int i:ids) processRow(i,v); return v; } @Override public double [] getDoubles(double [] vals, int from, int to, double NA){ for(int i = from; i < to; i++) { long x = UnsafeUtils.get8(_mem, 8*i); vals[i-from] = (x == _NA)?NA:x; } return vals; } @Override public double [] getDoubles(double [] vals, int [] ids){ int k = 0; for(int i:ids) { long x = UnsafeUtils.get8(_mem, 8*i); vals[k++] = (x == _NA)?Double.NaN:x; } return vals; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C8DChunk.java
package water.fvec; import water.util.UnsafeUtils; /** * The empty-compression function, where data is in 'double's. */ public class C8DChunk extends Chunk { C8DChunk( byte[] bs ) { _mem=bs; _start = -1; set_len(_mem.length>>3); } @Override protected final long at8_impl( int i ) { double res = UnsafeUtils.get8d(_mem, i << 3); if( Double.isNaN(res) ) throw new IllegalArgumentException("at8_abs but value is missing"); return (long)res; } @Override protected final double atd_impl( int i ) { return UnsafeUtils.get8d(_mem,i<<3) ; } @Override protected final boolean isNA_impl( int i ) { return Double.isNaN(UnsafeUtils.get8d(_mem,i<<3)); } @Override boolean set_impl(int idx, long l) { return false; } /** * Fast explicit set for double. * @param i * @param d */ public void set8D(int i, double d) {UnsafeUtils.set8d(_mem,i<<3,d);} public double get8D(int i) {return UnsafeUtils.get8d(_mem,i<<3);} @Override boolean set_impl(int i, double d) { UnsafeUtils.set8d(_mem,i<<3,d); return true; } @Override boolean set_impl(int i, float f ) { UnsafeUtils.set8d(_mem,i<<3,f); return true; } @Override boolean setNA_impl(int idx) { UnsafeUtils.set8d(_mem,(idx<<3),Double.NaN); return true; } // 3.3333333e33 // public int pformat_len0() { return 22; } // public String pformat0() { return "% 21.15e"; } @Override public final void initFromBytes () { _start = -1; _cidx = -1; set_len(_mem.length>>3); assert _mem.length == _len <<3; } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to) { for(int i = from; i < to; i++) v.addValue(UnsafeUtils.get8d(_mem,8*i)); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int[] ids) { for(int i:ids) v.addValue(UnsafeUtils.get8d(_mem,8*i)); return v; } @Override public double [] getDoubles(double [] vals, int from, int to, double NA){ for(int i = from; i < to; i++) { double d = UnsafeUtils.get8d(_mem, 8 * i); vals[i - from] = Double.isNaN(d)?NA:d; } return vals; } @Override public double [] getDoubles(double [] vals, int [] ids){ int k = 0; for(int i:ids) vals[k++] = UnsafeUtils.get8d(_mem,8*i); return vals; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/C8DVolatileChunk.java
package water.fvec; import water.*; import water.util.UnsafeUtils; /** * The empty-compression function, where data is in 'double's. * Can only be used locally (intentionally does not serialize). * Intended for temporary data which gets modified frequently. * Exposes data directly as double[] */ public final class C8DVolatileChunk extends Chunk { private transient double [] _ds; C8DVolatileChunk(double[] ds ) {_start = -1; _len = ds.length; _ds = ds; } public double [] getValues(){return _ds;} @Override protected final long at8_impl( int i ) { double res = _ds[i]; if( Double.isNaN(res) ) throw new IllegalArgumentException("at8_abs but value is missing"); return (long)res; } @Override protected final double atd_impl( int i ) { return _ds[i] ; } @Override protected final boolean isNA_impl( int i ) { return Double.isNaN(_ds[i]); } @Override boolean set_impl(int idx, long l) { double d = l; if(d != l) return false; _ds[idx] = d; return true; } @Override boolean set_impl(int i, double d) { _ds[i] = d; return true; } @Override boolean set_impl(int i, float f ) { _ds[i] = f; return true; } public boolean isVolatile() {return true;} @Override boolean setNA_impl(int idx) { UnsafeUtils.set8d(_mem,(idx<<3),Double.NaN); return true; } @Override public final void initFromBytes () { _len = _mem.length >> 3; _ds = MemoryManager.malloc8d(_len); for(int i = 0; i < _ds.length; ++i) _ds[i] = UnsafeUtils.get8d(_mem,8*i); _mem = null; } @Override public byte [] asBytes() { byte [] res = MemoryManager.malloc1(_len*8); for(int i = 0; i < _len; ++i) UnsafeUtils.set8d(res,8*i,_ds[i]); return res; } @Override public Futures close( int cidx, Futures fs ) { if(chk2() != null) return chk2().close(cidx,fs); Value v = new Value(_vec.chunkKey(cidx),this,_len*8,Value.ICE); DKV.put(v._key,v,fs); return fs; } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to) { for(int i = from; i < to; i++) v.addValue(_ds[i]); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int[] ids) { for(int i:ids) v.addValue(_ds[i]); return v; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/CBSChunk.java
package water.fvec; import water.MemoryManager; /** A simple chunk for boolean values. In fact simple bit vector. * Each boolean is represented by 2bits since we need to represent NA. */ public class CBSChunk extends Chunk { static protected final byte _NA = 0x02; // Internal representation of NA static protected final int _OFF = 2; private transient byte _bpv; public byte bpv() { return _bpv; } //bits per value private transient byte _gap;// number of trailing unused bits in the end (== _len % 8, we allocate bytes, but our length i generally not multiple of 8) public byte gap() { return _gap; } //number of trailing unused bits in the end public CBSChunk(boolean [] vals) { _len = vals.length; _bpv = 1; int memlen = _len >> 3 + (_len & 7) == 0?0:1; _mem = MemoryManager.malloc1(memlen); _mem[0] = _gap = (byte)(vals.length & 7); _mem[1] = _bpv; for(int i = 0; i < vals.length; ++i) if(vals[i])write(i,(byte)1); } public CBSChunk(byte[] bs) { _mem = bs; initFromBytes(); } public CBSChunk(int len, int bpv) { _gap = (byte) ((8 - (len*bpv & 7)) & 7); int clen = CBSChunk._OFF + (len >> (3 - bpv + 1)) + (_gap == 0?0:1); byte [] bs = MemoryManager.malloc1(clen); bs[0] = _gap; bs[1] = _bpv = (byte)bpv; assert ((clen - _OFF) - (_gap == 0?0:1) == (len >> (3-bpv+1))); _mem = bs; _start = -1; _len = len; } @Override protected long at8_impl(int idx) { byte b = read(idx); if( b == _NA ) throw new IllegalArgumentException("at8_abs but value is missing"); return b; } @Override protected double atd_impl(int idx) { byte b = read(idx); return b == _NA ? Double.NaN : b; } @Override protected final boolean isNA_impl( int i ) { return read(i)==_NA; } private void set_byte(int idx, byte val){ int bix = _OFF + ((idx*_bpv)>>3); // byte index int off = _bpv*idx & 7; // index within the byte int mask = ~((1 | _bpv) << off); _mem[bix] = (byte)((_mem[bix] & mask) | (val << off)); // 1 or 3 for 1bit per value or 2 bits per value } void write(int idx, byte val){ int bix = _OFF + ((idx*_bpv)>>3); // byte index int off = _bpv*idx & 7; // index within the byte write(bix, off,val); } protected byte read(int idx) { int bix = _OFF + ((idx*_bpv)>>3); // byte index int off = _bpv*idx & 7; // index within the byte int mask = (1 | _bpv); // 1 or 3 for 1bit per value or 2 bits per value return read(_mem[bix], off,mask); } @Override boolean set_impl(int idx, long l) { if (l == 1 || l == 0) { set_byte(idx, (byte)l); return true; } return false; } @Override boolean set_impl(int idx, double d) { if(Double.isNaN(d)) return setNA_impl(idx); if(d == 0 || d == 1) { set_byte(idx,(byte)d); return true; } return false; } @Override boolean set_impl(int idx, float f ) { if(Float.isNaN(f)) return setNA_impl(idx); if(f == 0 || f == 1) { set_byte(idx,(byte)f); return true; } return false; } @Override boolean setNA_impl(int idx) { if(_bpv == 2) { set_byte(idx, _NA); return true; } return false; } private void processRow(int r, ChunkVisitor v){ int i = read(r); if(i == _NA) v.addNAs(1); else v.addValue(i); } @Override public ChunkVisitor processRows(ChunkVisitor v, int from, int to){ for(int i = from; i < to; ++i) processRow(i,v); return v; } @Override public ChunkVisitor processRows(ChunkVisitor v, int... rows){ for(int i:rows) processRow(i,v); return v; } // /** Writes 1bit from value into b at given offset and return b */ // public static byte write1b(byte b, byte val, int off) { // val = (byte) ((val & 0x1) << (7-off)); // return (byte) (b | val); // } // /** Writes 2bits from value into b at given offset and return b */ // public static byte write2b(byte b, byte val, int off) { // val = (byte) ((val & 0x3) << (6-off)); // 0000 00xx << (6-off) // return (byte) (b | val); // } private byte read(int b, int off, int mask){ return (byte)((b >> off) & mask); } private byte write(int bix, int off, int val){ return _mem[bix] |= (val << off); } // /** Reads 1bit from given b in given offset. */ // public static byte read1b(byte b, int off) { return (byte) ((b >> (7-off)) & 0x1); } // /** Reads 2bit from given b in given offset. */ // public static byte read2b(byte b, int off) { return (byte) ((b >> (6-off)) & 0x3); } /** Returns compressed len of the given array length if the value if represented by bpv-bits. */ public static int clen(int values, int bpv) { int len = (values*bpv) >> 3; return values*bpv % 8 == 0 ? len : len + 1; } @Override double min() { return 0; } @Override double max() { return 1; } @Override protected final void initFromBytes () { _start = -1; _cidx = -1; _gap = _mem[0]; _bpv = _mem[1]; set_len(((_mem.length - _OFF)*8 - _gap) / _bpv); } @Override public boolean hasFloat() {return false;} }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/CSChunk.java
package water.fvec; import water.util.PrettyPrint; import water.util.UnsafeUtils; /** * Created by tomas on 8/14/17. * * Chunk storing 1/2/4 byte values with an offset and/or scale. * Used for fixed point decimal numbers or scaled/offseted integers. * * value(i) = (stored_bytes(i) + bias)*scale; if scale > 1 or scale == 1/x and (stored_bytes(i) + bias)*(1/x) == (stored_bytes(i) + bias)/x for all i-s * or * value(i) = (stored_bytes(i) + bias)/scale ; otherwise * */ public abstract class CSChunk extends Chunk { static protected final int _OFF=8+4+4; private transient double _scale; private transient long _bias; private transient boolean _isDecimal; CSChunk( byte[] bs, long bias, int scale, int szLog) { _mem = bs; _start = -1; set_len((_mem.length - _OFF) >> szLog); _bias = bias; UnsafeUtils.set8(_mem, 0, bias); UnsafeUtils.set4(_mem, 8, scale); _scale = PrettyPrint.pow10(1,scale); UnsafeUtils.set4(_mem,12,szLog); } protected void setDecimal(){ _isDecimal = true; _scale = PrettyPrint.pow10(1,-UnsafeUtils.get4(_mem,8)); UnsafeUtils.set4(_mem,12,-UnsafeUtils.get4(_mem,12)-1); } private int getSzLog(){ int x = UnsafeUtils.get4(_mem,12); return x < 0?-x-1:x; } public final double scale() { return _isDecimal?1.0/_scale:_scale; } @Override public final byte precision() { return (byte)Math.max(UnsafeUtils.get4(_mem,8),0); } protected final double getD(int x, int NA){return getD(x,NA,Double.NaN);} protected final double getD(int x, int NA, double naImpute){ return x == NA?naImpute:_isDecimal?(_bias + x)/_scale:(_bias + x)*_scale; } protected final long get8(int x) { return (_bias + x)*(long)(_scale); } @Override public final boolean hasFloat(){ return _isDecimal || _scale < 1; } @Override public final void initFromBytes () { _start = -1; _cidx = -1; set_len((_mem.length-_OFF) >> getSzLog()); _bias = UnsafeUtils.get8 (_mem,0); int x = UnsafeUtils.get4(_mem,8); int szLog = UnsafeUtils.get4(_mem,12); _isDecimal = szLog < 0; _scale = PrettyPrint.pow10(1,_isDecimal?-x:x); } @Override protected long at8_impl( int i ) { double res = atd_impl(i); // note: |mantissa| <= 4B => double is ok if(Double.isNaN(res)) throw new IllegalArgumentException("at8_abs but value is missing"); return (long)res; } @Override public final boolean set_impl(int idx, long l) { double d = (double)l; if(d != l) return false; return set_impl(idx,d); } @Override public final boolean set_impl(int idx, float f) { return set_impl(idx,(double)f); } protected final int getScaledValue(double d, int NA){ assert !Double.isNaN(d):"NaN should be handled separately"; return (int)((_isDecimal?d*_scale:(d/_scale))-_bias); } @Override public final <T extends ChunkVisitor> T processRows(T v, int from, int to) { if(v.expandedVals()){ processRows2(v,from,to,_bias,UnsafeUtils.get4(_mem,8)); } else processRows2(v,from,to); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int[] ids) { if(v.expandedVals()){ processRows2(v,ids,_bias,UnsafeUtils.get4(_mem,8)); } else processRows2(v,ids); return v; } protected abstract <T extends ChunkVisitor> T processRows2(T v, int from, int to, long bias, int exp) ; protected abstract <T extends ChunkVisitor> T processRows2(T v, int from, int to); protected abstract <T extends ChunkVisitor> T processRows2(T v, int [] ids, long bias, int exp) ; protected abstract <T extends ChunkVisitor> T processRows2(T v, int [] ids); }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/CStrChunk.java
package water.fvec; import water.*; import water.util.SetOfBytes; import water.util.StringUtils; import water.util.UnsafeUtils; import water.parser.BufferedString; import java.util.Arrays; /** * The empty-compression function, where data is in 'string's. */ public class CStrChunk extends Chunk { static final int NA = -1; static protected final int _OFF=4+1; private int _valstart; public boolean _isAllASCII = false; public CStrChunk() {} /** * Empty-compression function, where data is a constant string * @param s Constant string * @param len Chunk length */ public CStrChunk(String s, int len){ byte[] sBytes = StringUtils.bytesOf(s); sBytes = Arrays.copyOf(sBytes,sBytes.length+1); sBytes[sBytes.length-1] = 0; init(sBytes.length, StringUtils.bytesOf(s),len,len,null,null); } /** * Empty-compression function, where data is in 'string's. * @param sslen Next offset into ss for placing next String * @param ss Bytes of appended strings, including trailing 0 * @param sparseLen Length of sparse chunk (number of extracted (non-zero) elements) * @param idxLen Length of chunk * @param id Indices (row numbers) of stored values, used for sparse * @param is Index of strings - holds offsets into ss[]. is[i] == -1 means NA/sparse */ public CStrChunk(int sslen, byte[] ss, int sparseLen, int idxLen, int[] id, int[] is) { init(sslen,ss,sparseLen,idxLen,id,is); } private void init (int sslen, byte[] ss, int sparseLen, int idxLen, int[] id, int[] is) { _start = -1; _valstart = idx(idxLen); _len = idxLen; _mem = MemoryManager.malloc1(_valstart + sslen, false); UnsafeUtils.set4(_mem, 0, _valstart); // location of start of strings Arrays.fill(_mem,_OFF,_valstart,(byte)-1); // Indicate All Is NA's for( int i = 0; i < sparseLen; ++i ) // Copy the sparse indices UnsafeUtils.set4(_mem, idx(id==null ? i : id[i]), is==null ? 0 : is[i]); //Need to check if id and is are null since both are not always needed for mem allocation UnsafeUtils.copyMemory(ss,0,_mem,_valstart,sslen); _isAllASCII = true; for(int i = _valstart; i < _mem.length; ++i) { byte c = _mem[i]; if ((c & 0x80) == 128) { //value beyond std ASCII _isAllASCII = false; break; } } UnsafeUtils.set1(_mem, 4, (byte) (_isAllASCII ? 1 : 0)); // isAllASCII flag } private int idx(int i) { return _OFF+(i<<2); } @Override public boolean setNA_impl(int idx) { return false; } @Override public boolean set_impl(int idx, float f) { if (Float.isNaN(f)) return false; else throw new IllegalArgumentException("Operation not allowed on string vector.");} @Override public boolean set_impl(int idx, double d) { if (Double.isNaN(d)) return false; else throw new IllegalArgumentException("Operation not allowed on string vector.");} @Override public boolean set_impl(int idx, long l) { throw new IllegalArgumentException("Operation not allowed on string vector.");} @Override public boolean set_impl(int idx, String str) { return false; } @Override public boolean isNA_impl(int idx) { int off = intAt(idx); return off == NA; } public int intAt(int i) { return UnsafeUtils.get4(_mem, idx(i)); } public byte byteAt(int i) { return _mem[_valstart+i]; } public int lengthAtOffset(int off) { int len = 0; while (byteAt(off + len) != 0) len++; return len; } @Override public long at8_impl(int idx) { throw new IllegalArgumentException("Operation not allowed on string vector.");} @Override public double atd_impl(int idx) { throw new IllegalArgumentException("Operation not allowed on string vector.");} @Override public BufferedString atStr_impl(BufferedString bStr, int idx) { int off = intAt(idx); if( off == NA ) return null; int len = lengthAtOffset(off); assert len >= 0 : getClass().getSimpleName() + ".atStr_impl: len=" + len + ", idx=" + idx + ", off=" + off; return bStr.set(_mem,_valstart+off,len); } @Override protected final void initFromBytes () { _start = -1; _cidx = -1; _valstart = UnsafeUtils.get4(_mem, 0); byte b = UnsafeUtils.get1(_mem,4); _isAllASCII = b != 0; set_len((_valstart-_OFF)>>2); } @Override public ChunkVisitor processRows(ChunkVisitor nc, int from, int to){ BufferedString bs = new BufferedString(); for(int i = from; i < to; i++) nc.addValue(atStr(bs,i)); return nc; } @Override public ChunkVisitor processRows(ChunkVisitor nc, int... rows){ BufferedString bs = new BufferedString(); for(int i:rows) nc.addValue(atStr(bs,i)); return nc; } /** * Optimized toLower() method to operate across the entire CStrChunk buffer in one pass. * This method only changes the values of ASCII uppercase letters in the text. * * NewChunk is the same size as the original. * * @param nc NewChunk to be filled with the toLower version of ASCII strings in this chunk * @return Filled NewChunk */ public NewChunk asciiToLower(NewChunk nc) { // copy existing data nc = this.extractRows(nc, 0,_len); //update offsets and byte array for(int i= 0; i < nc._sslen; i++) { if (nc._ss[i] > 0x40 && nc._ss[i] < 0x5B) // check for capital letter nc._ss[i] += 0x20; // lower it } return nc; } /** * Optimized toUpper() method to operate across the entire CStrChunk buffer in one pass. * This method only changes the values of ASCII lowercase letters in the text. * * NewChunk is the same size as the original. * * @param nc NewChunk to be filled with the toUpper version of ASCII strings in this chunk * @return Filled NewChunk */ public NewChunk asciiToUpper(NewChunk nc) { // copy existing data nc = this.extractRows(nc, 0,_len); //update offsets and byte array for(int i= 0; i < nc._sslen; i++) { if (nc._ss[i] > 0x60 && nc._ss[i] < 0x7B) // check for capital letter nc._ss[i] -= 0x20; // upper it } return nc; } /** * Optimized trim() method to operate across the entire CStrChunk buffer in one pass. * This mimics Java String.trim() by only considering characters of value * <code>'&#92;u0020'</code> or less as whitespace to be trimmed. This means that like * Java's String.trim() it ignores 16 of the 17 characters regarded as a space in UTF. * * NewChunk is the same size as the original, despite trimming. * * @param nc NewChunk to be filled with trimmed version of strings in this chunk * @return Filled NewChunk */ public NewChunk asciiTrim(NewChunk nc) { // copy existing data nc = this.extractRows(nc, 0,_len); //update offsets and byte array for(int i=0; i < _len; i++) { int j = 0; int off = UnsafeUtils.get4(_mem,idx(i)); if (off != NA) { //UTF chars will appear as negative values. In Java spec, space is any char 0x20 and lower while( _mem[_valstart+off+j] > 0 && _mem[_valstart+off+j] < 0x21) j++; if (j > 0) nc.set_is(i,off + j); while( _mem[_valstart+off+j] != 0 ) j++; //Find end j--; while( _mem[_valstart+off+j] > 0 && _mem[_valstart+off+j] < 0x21) { //March back to find first non-space nc._ss[off+j] = 0; //Set new end j--; } } } return nc; } /** * Optimized substring() method for a buffer of only ASCII characters. * The presence of UTF-8 multi-byte characters would give incorrect results * for the string length, which is required here. * * @param nc NewChunk to be filled with substrings in this chunk * @param startIndex The beginning index of the substring, inclusive * @param endIndex The ending index of the substring, exclusive * @return Filled NewChunk */ public NewChunk asciiSubstring(NewChunk nc, int startIndex, int endIndex) { // copy existing data nc = this.extractRows(nc, 0,_len); //update offsets and byte array for (int i = 0; i < _len; i++) { int off = UnsafeUtils.get4(_mem, idx(i)); if (off != NA) { int len = 0; while (_mem[_valstart + off + len] != 0) len++; //Find length nc.set_is(i,startIndex < len ? off + startIndex : off + len); for (; len > endIndex - 1; len--) { nc._ss[off + len] = 0; //Set new end } } } return nc; } /** * Optimized length() method for a buffer of only ASCII characters. * This is a straight byte count for each word in the chunk. The presence * of UTF-8 multi-byte characters would give incorrect results. * * @param nc NewChunk to be filled with lengths of strings in this chunk * @return Filled NewChunk */ public NewChunk asciiLength(NewChunk nc) { //pre-allocate since size is known nc.alloc_mantissa(_len); nc.alloc_exponent(_len); // sadly, a waste // fill in lengths for(int i=0; i < _len; i++) { int off = UnsafeUtils.get4(_mem,idx(i)); int len = 0; if (off != NA) { while (_mem[_valstart + off + len] != 0) len++; nc.addNum(len, 0); } else nc.addNA(); } return nc; } public NewChunk asciiEntropy(NewChunk nc) { nc.alloc_doubles(_len); for (int i = 0; i < _len; i++) { double entropy = entropyAt(i); if (Double.isNaN(entropy)) nc.addNA(); else nc.addNum(entropy); } return nc; } double entropyAt(int i) { int off = intAt(i); if (off == NA) return Double.NaN; int[] frq = new int[256]; int len = lengthAtOffset(off); for (int j = 0; j < len; j++) { frq[0xff & byteAt(off + j)]++; } double sum = 0; for (int b = 0; b < 256; b++) { int f = frq[b]; if (f > 0) { double x = (double)f / len; sum += x * Math.log(x); } } return - sum / Math.log(2); } /** * Optimized lstrip() & rstrip() methods to operate across the entire CStrChunk buffer in one pass. * * NewChunk is the same size as the original, despite trimming. * * @param nc NewChunk to be filled with strip version of strings in this chunk * @param chars chars to strip, treated as ASCII * @return Filled NewChunk */ public NewChunk asciiLStrip(NewChunk nc, String chars) { SetOfBytes set = new SetOfBytes(chars); //update offsets and byte array for(int i=0; i < _len; i++) { int off = intAt(i); if (off != NA) { while (set.contains(byteAt(off))) off++; int len = lengthAtOffset(off); nc.addStr(new BufferedString(_mem, _valstart+off, len)); } else nc.addNA(); } return nc; } public NewChunk asciiRStrip(NewChunk nc, String chars) { SetOfBytes set = new SetOfBytes(chars); //update offsets and byte array for(int i=0; i < _len; i++) { int off = intAt(i); if (off != NA) { int pos = off + lengthAtOffset(off); while (pos --> off && set.contains(byteAt(pos))); nc.addStr(new BufferedString(_mem, _valstart+off, pos - off + 1)); } else nc.addNA(); } return nc; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/CUDChunk.java
package water.fvec; import water.MemoryManager; import water.util.UnsafeUtils; import java.util.HashMap; import java.util.Map; /** * The "few unique doubles"-compression function */ public class CUDChunk extends Chunk { public static int MAX_UNIQUES=256; public static int computeByteSize(int uniques, int len) { return 4 + 4 // _len + numUniques + (uniques << 3) //unique double values + (len << 1); //mapping of row -> unique value index (0...255) } int numUniques; CUDChunk() {} CUDChunk(byte[] bs, HashMap<Long,Byte> hs, int len) { _start = -1; numUniques = hs.size(); set_len(len); _mem = MemoryManager.malloc1(computeByteSize(numUniques, _len), false); UnsafeUtils.set4(_mem, 0, _len); UnsafeUtils.set4(_mem, 4, numUniques); int j=0; //create the mapping and also store the unique values (as longs) for (Map.Entry<Long,Byte> e : hs.entrySet()) { e.setValue((byte) (j - 128)); //j is in 0...256 -> byte value needs to be in -128...127 for storage UnsafeUtils.set8(_mem, 8 + (j << 3), e.getKey()); j++; } // store the mapping for (int i=0; i<len; ++i) UnsafeUtils.set1(_mem, 8 + (numUniques << 3) + i, hs.get(Double.doubleToLongBits(UnsafeUtils.get8d(bs, i << 3)))); } @Override protected final long at8_impl( int i ) { double res = atd_impl(i); if( Double.isNaN(res) ) throw new IllegalArgumentException("at8_impl but value is missing"); return (long)res; } @Override protected final double atd_impl( int i ) { int whichUnique = (UnsafeUtils.get1(_mem, 8 + (numUniques << 3) + i)+128); return Double.longBitsToDouble(UnsafeUtils.get8(_mem, 8 + (whichUnique << 3))); } @Override public double [] getDoubles(double [] vals, int from, int to) { return getDoubles(vals,from,to,Double.NaN); } @Override public double [] getDoubles(double [] vals, int from, int to, double NA) { double [] uniques = new double[numUniques]; for(int i = 0; i < numUniques; ++i) { uniques[i] = Double.longBitsToDouble(UnsafeUtils.get8(_mem, 8 + (i << 3))); if(Double.isNaN(uniques[i])) uniques[i] = NA; } for(int i = 0; i < _len; ++i) vals[i] = uniques[(UnsafeUtils.get1(_mem, 8 + (numUniques << 3) + i)+128)]; return vals; } @Override protected final boolean isNA_impl( int i ) { return Double.isNaN(atd_impl(i)); } @Override boolean set_impl(int idx, long l) { return false; } @Override boolean set_impl(int i, double d) { for (int j = 0; j < numUniques; ++j) { if (Double.compare(Double.doubleToLongBits(d), UnsafeUtils.get8(_mem, 8 + (j << 3))) == 0) { UnsafeUtils.set1(_mem, 8 + (numUniques << 3) + i, (byte) (j-128)); return true; } } return false; } @Override boolean set_impl(int i, float f ) { return set_impl(i, (double)f); } @Override boolean setNA_impl(int idx) { return set_impl(idx, Double.NaN); } @Override public ChunkVisitor processRows(ChunkVisitor nc, int from, int to){ for(int i = from; i < to; i++) nc.addValue(atd(i)); return nc; } @Override public ChunkVisitor processRows(ChunkVisitor nc, int... rows){ for(int i:rows) nc.addValue(atd(i)); return nc; } @Override protected final void initFromBytes () { _start = -1; _cidx = -1; _len = UnsafeUtils.get4(_mem, 0); numUniques = UnsafeUtils.get4(_mem, 4); set_len(_len); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/CXFChunk.java
package water.fvec; import water.H2O; import water.util.UnsafeUtils; // Sparse chunk. public class CXFChunk extends CXIChunk { protected CXFChunk(byte [] mem){ super(mem); } private double getVal(int x){ switch(_elem_sz) { case 8: return UnsafeUtils.get4f(_mem, x + 4); case 12: return UnsafeUtils.get8d(_mem, x + 4); default: throw H2O.unimpl(); } } @Override public long at8_impl(int idx){ int x = findOffset(idx); if(x < 0) { if(_isNA) throw new RuntimeException("at4 but the value is missing!"); return 0; } double val = getVal(x); if(Double.isNaN(val)) throw new RuntimeException("at4 but the value is missing!"); return (long)val; } @Override public double atd_impl(int idx) { int x = findOffset(idx); if(x < 0) return _isNA?Double.NaN:0; return getVal(x); } @Override public Chunk deepCopy() {return new CXFChunk(_mem.clone());} @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to){ int prevId = from-1; int x = from == 0?_OFF: findOffset(from); if(x < 0) x = -x-1; while(x < _mem.length){ int id = getId(x); if(id >= to)break; if(_isNA) v.addNAs(id-prevId-1); else v.addZeros(id-prevId-1); v.addValue(getVal(x)); prevId = id; x+=_elem_sz; } if(_isNA) v.addNAs(to-prevId-1); else v.addZeros(to-prevId-1); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int [] ids){ int x = _OFF; int k = 0; int zeros = 0; while(k < ids.length) { int idk = ids[k]; assert ids[k] >= 0 && (k == 0 || ids[k] > ids[k-1]); int idx = ids[ids.length-1]+1; while(x < _mem.length && (idx = getId(x)) < idk) x += _elem_sz; if(x == _mem.length){ zeros += ids.length - k; break; } if(idx == idk){ if(_isNA) v.addNAs(zeros); else v.addZeros(zeros); v.addValue(getVal(x)); zeros = 0; x+=_elem_sz; } else zeros++; k++; } if(zeros > 0){ if(_isNA) v.addNAs(zeros); else v.addZeros(zeros); } return v; } @Override public boolean hasFloat(){return true;} }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/CXIChunk.java
package water.fvec; import water.H2O; import water.util.UnsafeUtils; // Sparse chunk. public class CXIChunk extends Chunk { private static long [] _NAS = new long[]{-1/* not used, binary chunks can't have NAs */,/* not used*/-1,C2Chunk._NA,-1,C4Chunk._NA,-1,-1,-1,C8Chunk._NA}; public static long NA(int val_sz){return _NAS[val_sz];} transient boolean _isNA; // na sparse or zero sparse transient int _val_sz; transient int _elem_sz; // transient int _elemsz_log; // 2 or 3 for 4 or 8 byte per element static final int _OFF = 8; // 4B of len, 1B id sz 1B value, 1B isNA, 1B padding protected CXIChunk(byte [] mem){ _mem = mem; initFromBytes(); } @Override public int sparseLenZero(){ return isSparseZero()?sparseLen():_len; } @Override public int sparseLenNA(){ return isSparseNA()?sparseLen():_len; } protected final int getId(int x){ if(x == _mem.length) return _len; int id_sz = _elem_sz - _val_sz; return id_sz == 4?UnsafeUtils.get4(_mem,x):0xFFFF&UnsafeUtils.get2(_mem,x); } private long getVal(int x){ switch(_val_sz){ case 0: return 1; case 2: return UnsafeUtils.get2(_mem,x+2); case 4: return UnsafeUtils.get4(_mem,x+4); case 8: return UnsafeUtils.get8(_mem,x+4); default: throw H2O.unimpl(); } } private double getFVal(int x){ long ival = getVal(x); return ival == _NAS[_val_sz]?Double.NaN:ival; } @Override public final boolean isSparseNA(){return _isNA;} @Override public final boolean isSparseZero(){return !_isNA;} @Override protected final void initFromBytes() { _start = -1; _cidx = -1; _len = UnsafeUtils.get4(_mem,0); int id_sz = _mem[4]&0xFF; _val_sz = _mem[5]&0xFF; _elem_sz = _val_sz + id_sz; _isNA = (0xFF&_mem[6]) == 1; _previousOffset = _OFF; } protected final int sparseLen(){ return (_mem.length - _OFF) / _elem_sz; } protected final int getOff(int id){return _OFF + _elem_sz*id;} protected final int getIdx(int off){return (off-_OFF)/_elem_sz;} protected final int findOffset(int i) { // do binary search int off = _previousOffset; int id = getId(off); if(id == i) return off; if(id < i && (id = getId(off+=_elem_sz)) == i) { _previousOffset = off; return off; } int lb = id < i?getIdx(off):0; int ub = id > i?getIdx(off):sparseLen(); while (lb < ub) { int mid = lb + ((ub - lb) >> 1); off = getOff(mid); int x = getId(off); if (x == i) { _previousOffset = off; return off; } if (x < i) lb = mid + 1; else ub = mid; } return -getOff(ub)-1; } @Override public long at8_impl(int idx){ int x = findOffset(idx); if(x < 0) { if(_isNA) throw new RuntimeException("at8 but the value is missing!"); return 0; } long val = getVal(x); if(val == _NAS[_val_sz]) throw new RuntimeException("at4 but the value is missing!"); return val; } @Override public double atd_impl(int idx) { int x = findOffset(idx); if(x < 0) return _isNA?Double.NaN:0; return getFVal(x); } @Override public final boolean isNA_impl( int i ) {return Double.isNaN(atd(i));} @Override boolean set_impl(int idx, long l) { return false; } @Override boolean set_impl(int idx, double d) { return false; } @Override boolean set_impl(int idx, float f) { return false; } @Override boolean setNA_impl(int idx) { return false; } @Override public boolean hasNA() { return true; } @Override public Chunk deepCopy() {return new CXIChunk(_mem.clone());} public final int len(){return _len;} private transient int _previousOffset = _OFF; @Override public final int nextNZ(int i){ int x = findOffset(i); if(x < 0) x = -x-1-_elem_sz; _previousOffset = x += _elem_sz; return getId(x); } @Override public final int nextNZ(int rid, boolean onlyTrueZero) { return onlyTrueZero && _isNA ? rid + 1 : nextNZ(rid); } @Override public <T extends ChunkVisitor> T processRows(T v, int from, int to){ int prevId = from-1; int x = from == 0?_OFF: findOffset(from); if(x < 0) x = -x-1; while(x < _mem.length){ int id = getId(x); if(id >= to)break; if(_isNA) v.addNAs(id-prevId-1); else v.addZeros(id-prevId-1); long val = getVal(x); if(val ==_NAS[_val_sz]) v.addNAs(1); else v.addValue(val); prevId = id; x+=_elem_sz; } if(_isNA) v.addNAs(to-prevId-1); else v.addZeros(to-prevId-1); return v; } @Override public <T extends ChunkVisitor> T processRows(T v, int [] ids){ int x = _OFF; int k = 0; int zeros = 0; while(k < ids.length) { int idk = ids[k]; assert ids[k] >= 0 && (k == 0 || ids[k] > ids[k-1]); int idx = ids[ids.length-1]+1; while(x < _mem.length && (idx = getId(x)) < idk) x += _elem_sz; if(x == _mem.length){ zeros += ids.length - k; break; } if(idx == idk){ if(_isNA) v.addNAs(zeros); else v.addZeros(zeros); long val = getVal(x); if(val == _NAS[_val_sz]) v.addNAs(1); else v.addValue(val); zeros = 0; x+=_elem_sz; } else zeros++; k++; } if(zeros > 0){ if(_isNA) v.addNAs(zeros); else v.addZeros(zeros); } return v; } @Override public boolean hasFloat(){return false;} }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/CategoricalWrappedVec.java
package water.fvec; import water.AutoBuffer; import water.Key; import water.DKV; import water.MRTask; import water.util.ArrayUtils; import java.util.Arrays; import java.util.HashMap; /** A vector transforming values of given vector according to given domain * mapping - currently only used to transform categorical columns but in theory would * work for any dense-packed Int column. Expected usage is to map from a new * dataset to the domain-mapping expected by a model (which will match the * dataset it was trained on). * * <p>The Vector's domain is the union of the Test and Train domains. * * <p>The mapping is defined by int[] array, size is input Test.domain.length. * Contents refer to values in the Train.domain. Extra values in the Test * domain are sorted after the Train.domain - so mapped values have to be * range-checked (note that returning some flag for NA, say -1, would also * need to be checked for). */ public class CategoricalWrappedVec extends WrappedVec { /** List of values from underlying vector which this vector map to a new * value in the union domain. */ int[] _map; int _p=0; /** Main constructor: convert from one categorical to another */ public CategoricalWrappedVec(Key key, int rowLayout, String[] toDomain, Key masterVecKey) { super(key, rowLayout, masterVecKey); computeMap(masterVec().domain(),toDomain,masterVec().isBad()); DKV.put(this); } /** Constructor just to generate the map and domain; used in tests or when * mixing categorical columns */ private CategoricalWrappedVec(Key key) { super(key, ESPC.rowLayout(key, new long[]{0}), null, null); } private static CategoricalWrappedVec makeTransientVec(String[] from, String[] to) { Key key = Vec.newKey(); CategoricalWrappedVec tmp = new CategoricalWrappedVec(key); tmp.computeMap(from, to, false); return tmp; } public static int[] computeMap(String[] from, String[] to) { return makeTransientVec(from, to)._map; } /** * Updates Vec's domain in-place (instead of creating a wrapper Vec) * @param source source Vec * @param toDomain target domain * @return updated instance of Vec */ public static Vec updateDomain(Vec source, String[] toDomain) { CategoricalWrappedVec tv = makeTransientVec(source.domain(), toDomain); new RemapTask(tv).doAll(source); source.setDomain(tv.domain()); DKV.put(source); return source; } private static class RemapTask extends MRTask<RemapTask> { private final int[] _map; private final int _p; private RemapTask(CategoricalWrappedVec vec) { _map = vec._map; _p = vec._p; } @Override public void map(Chunk c) { Chunk wc = new CategoricalWrappedChunk(c, c._vec, _map, _p); assert wc._len == c._len; for (int i = 0; i < wc._len; i++) { if (c.isNA(i)) continue;; c.set(i, wc.at8(i)); } } } @Override public Chunk chunkForChunkIdx(int cidx) { return new CategoricalWrappedChunk(masterVec().chunkForChunkIdx(cidx), this); } /** Compute a mapping from the 'from' domain to the 'to' domain. Strings in * the 'from' domain not in the 'to' domain are mapped past the end of the * 'to' values. Strings in the 'to' domain not in the 'from' domain * simply do not appear in the mapping. The returned map is always the same * length as the 'from' domain. Its contents have values from both * domains; the resulting domain is as big as the largest value in the map, * and only has strings from the 'from' domain (which probably overlap * somewhat with the 'to' domain). * * <p> Example: from={"Blue","Red","Green"}, to={"Green","Yellow","Blue"}.<br> * "Yellow" does not appear in the 'from' domain; "Red" does not appear in the 'to' domain.<br> * Returned map is {2,3,0}.<br> * Map length matches the 'from' domain length.<br> * Largest value is 3, so the domain is size 4.<br> * Domain is: {"Green","Yellow","Blue","Red"}<br> * Extra values in the 'from' domain appear, in-order in the 'from' domain, at the end. * @return mapping */ private void computeMap( String[] from, String[] to, boolean fromIsBad ) { // Identity? Build the cheapo non-map if( from==to || Arrays.equals(from,to) ) { _map = ArrayUtils.seq(0,to.length); setDomain(to); return; } // The source Vec does not have a domain, hence is an integer column. The // to[] mapping has the set of unique numbers, we need to map from those // numbers to the index to the numbers. if( from==null) { setDomain(to); if( fromIsBad ) { _map = new int[0]; return; } int min = Integer.valueOf(to[0]); int max = Integer.valueOf(to[to.length-1]); Vec mvec = masterVec(); if( !(mvec.isInt() && mvec.min() >= min && mvec.max() <= max) ) throw new NumberFormatException(); // Unable to figure out a valid mapping // FIXME this is a bit of a hack to allow adapTo calls to play nice with negative ints in the domain... if( Integer.valueOf(to[0]) < 0 ) { _p=Math.max(0,max); _map = new int[(_p /*positive array of values*/) + (-1*min /*negative array of values*/) + 1 /*one more to store "max" value*/]; for(int i=0;i<to.length;++i) { int v = Integer.valueOf(to[i]); if( v < 0 ) v = -1*v+_p; _map[v] = i; } return; } _map = new int[max+1]; for( int i=0; i<to.length; i++ ) _map[Integer.valueOf(to[i])] = i; return; } // The desired result Vec does not have a domain, hence is a numeric // column. For classification of numbers, we did an original toCategoricalVec // wrapping the numeric values up as Strings for the classes. Unwind that, // converting numeric strings back to their original numbers. _map = new int[from.length]; if( to == null ) { for( int i=0; i<from.length; i++ ) _map[i] = Integer.valueOf(from[i]); return; } // Full string-to-string mapping HashMap<String,Integer> h = new HashMap<>(); for( int i=0; i<to.length; i++ ) h.put(to[i],i); String[] ss = to; int extra = to.length; int actualLen = extra; for( int j=0; j<from.length; j++ ) { Integer x = h.get(from[j]); if( x!=null ) _map[j] = x; else { _map[j] = extra++; if (extra > ss.length) { ss = Arrays.copyOf(ss, 2*ss.length); } ss[extra-1] = from[j]; actualLen = extra; } } setDomain(Arrays.copyOf(ss, actualLen)); } @Override public Vec doCopy() { return new CategoricalWrappedVec(group().addVec(),_rowLayout, domain(), _masterVecKey); } public static class CategoricalWrappedChunk extends Chunk { public final transient Chunk _c; // Test-set map final transient int[] _map; final transient int _p; CategoricalWrappedChunk(Chunk c, CategoricalWrappedVec vec) { this(c, vec, vec._map, vec._p); } private CategoricalWrappedChunk(Chunk c, Vec vec, int[] map, int p) { _c = c; set_len(_c._len); _start = _c._start; _vec = vec; _cidx = _c._cidx; _map = map; _p = p; } // Returns the mapped value. {@code _map} covers all the values in the // master Chunk, so no AIOOBE. Missing values in the master Chunk return // the usual NaN. @Override protected double atd_impl(int idx) { return _c.isNA_impl(idx) ? Double.NaN : at8_impl(idx); } // Returns the mapped value. {@code _map} covers all the values in the // master Chunk, so no AIOOBE. Missing values in the master Chunk throw // the normal missing-value exception when loading from the master. @Override protected long at8_impl(int idx) { int at8 = (int)_c.at8_impl(idx); if( at8 >= 0 ) return _map[at8]; else return _map[-1*at8+_p]; } // Returns true if the masterVec is missing, false otherwise @Override protected boolean isNA_impl(int idx) { return _c.isNA_impl(idx); } @Override boolean set_impl(int idx, long l) { return false; } @Override boolean set_impl(int idx, double d) { return false; } @Override boolean set_impl(int idx, float f) { return false; } @Override boolean setNA_impl(int idx) { return false; } @Override public ChunkVisitor processRows(ChunkVisitor nc, int from, int to){ for( int i=from; i< to; i++ ) if(isNA(i))nc.addNAs(1); else nc.addValue(at8(i)); return nc; } @Override public ChunkVisitor processRows(ChunkVisitor nc, int... rows){ for( int i:rows) if(isNA(i))nc.addNAs(1); else nc.addValue(at8(i)); return nc; } public static AutoBuffer write_impl(CategoricalWrappedVec v,AutoBuffer bb) { throw water.H2O.fail(); } @Override protected final void initFromBytes () { throw water.H2O.fail(); } @Override public boolean hasNA() { return _c.hasNA(); } @Override public Chunk deepCopy() { return extractRows(new NewChunk(this),0,_c._len).compress(); } } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/Chunk.java
package water.fvec; import water.*; import water.parser.BufferedString; import java.util.UUID; /** A compression scheme, over a chunk of data - a single array of bytes. * Chunks are mapped many-to-1 to a {@link Vec}. The <em>actual</em> vector * header info is in the Vec - which contains info to find all the bytes of * the distributed vector. Subclasses of this abstract class implement * (possibly empty) compression schemes. * * <p>Chunks are collections of elements, and support an array-like API. * Chunks are subsets of a Vec; while the elements in a Vec are numbered * starting at 0, any given Chunk has some (probably non-zero) starting row, * and a length which is smaller than the whole Vec. Chunks are limited to a * single Java byte array in a single JVM heap, and only an int's worth of * elements. Chunks support both the notions of a global row-number and a * chunk-local numbering. The global row-number calls are variants of {@code * at_abs} and {@code set_abs}. If the row is outside the current Chunk's * range, the data will be loaded by fetching from the correct Chunk. This * probably involves some network traffic, and if all rows are loaded then the * entire dataset will be pulled locally (possibly triggering an OutOfMemory). * * <p>The chunk-local numbering supports the common {@code for} loop iterator * pattern, using {@code at*} and {@code set} calls, and is faster than the * global row-numbering for tight loops (because it skips some range checks): * <pre>{@code * for (int row = 0; row < chunk._len; row++) * ...chunk.atd(row)... * }</pre> * * <p>The array-like API allows loading and storing elements in and out of * Chunks. When loading, values are decompressed. When storing, an attempt * to compress back into the actual underlying Chunk subclass is made; if this * fails the Chunk is "inflated" into a {@link NewChunk}, and the store * completed there. Later the NewChunk will be compressed (probably into a * different underlying Chunk subclass) and put back in the K/V store under * the same Key - effectively replacing the original Chunk; this is done when * {@link #close} is called, and is taken care of by the standard {@link * MRTask} calls. * * <p>Chunk updates are not multi-thread safe; the caller must do correct * synchronization. This is already handled by the Map/Reduce {MRTask) * framework. Chunk updates are not visible cross-cluster until the {@link * #close} is made; again this is handled by MRTask directly. * * <p>In addition to normal load and store operations, Chunks support the * notion a missing element via the {@link #isNA} call, and a "next non-zero" * notion for rapidly iterating over sparse data. * * <p><b>Data Types</b> * * <p>Chunks hold Java primitive values, timestamps, UUIDs, or Strings. All * the Chunks in a Vec hold the same type. Most of the types are compressed. * Integer types (boolean, byte, short, int, long) are always lossless. Float * and Double types might lose 1 or 2 ulps in the compression. Time data is * held as milliseconds since the Unix Epoch. UUIDs are held as 128-bit * integers (a pair of Java longs). Strings are compressed in various obvious * ways. Sparse data is held... sparsely; e.g. loading data in SVMLight * format will not "blow up" the in-memory representation. Categoricals/factors * are held as small integers, with a shared String lookup table on the side. * * <p>Chunks support the notion of <em>missing</em> data. Missing float and * double data is always treated as a NaN, both if read or written. There is * no equivalent of NaN for integer data; reading a missing integer value is a * coding error and will be flagged. If you are working with integer data * with missing elements, you must first check for a missing value before * loading it: * <pre>{@code * if( !chk.isNA(row) ) ...chk.at8(row).... * }</pre> * * <p>The same holds true for the other non-real types (timestamps, UUIDs, * Strings, or categoricals): they must be checked for missing before being * used. * * <p><b>Performance Concerns</b> * * <p>The standard {@code for} loop mentioned above is the fastest way to * access data; definitely faster (and less error prone) than iterating over * global row numbers. Iterating over a single Chunk is nearly always * memory-bandwidth bound. Often code will iterate over a number of Chunks * aligned together (the common use-case of looking a whole rows of a * dataset). Again, typically such a code pattern is memory-bandwidth bound * although the X86 will stop being able to prefetch well beyond 100 or 200 * Chunks. * * <p>Note that Chunk alignment is guaranteed within all the Vecs of a Frame: * Same numbered Chunks of <em>different</em> Vecs will have the same global * row numbering and the same length, enabling a particularly simple and * efficient way to iterate over all rows. * * <p>This example computes the Euclidean distance between all the columns and * a given point, and stores the squared distance back in the last column. * Note that due "NaN poisoning" if any row element is missing, the entire * distance calculated will be NaN. * <pre>{@code * final double[] _point; // The given point * public void map( Chunk[] chks ) { // Map over a set of same-numbered Chunks * for( int row=0; row < chks[0]._len; row++ ) { // For all rows * double dist=0; // Squared distance * for( int col=0; col < chks.length-1; col++ ) { // For all cols, except the last output col * double d = chks[col].atd(row) - _point[col]; // Distance along this dimension * dist += d*d; // Sum-squared-distance * } * chks[chks.length-1].set( row, dist ); // Store back the distance in the last col * } * }}</pre> */ public abstract class Chunk extends Iced<Chunk> implements Vec.Holder { public Chunk() {} private Chunk(byte [] bytes) {_mem = bytes;initFromBytes();} /** Global starting row for this local Chunk; a read-only field. */ transient long _start = -1; /** Global starting row for this local Chunk */ public final long start() { return _start; } /** Global index of this chunk filled during chunk load */ transient int _cidx = -1; /** Number of rows in this Chunk; publicly a read-only field. Odd API * design choice: public, not-final, read-only, NO-ACCESSOR. * * <p>NO-ACCESSOR: This is a high-performance field, and must have a known * zero-cost cost-model; accessors hide that cost model, and make it * not-obvious that a loop will be properly optimized or not. * * <p>not-final: set in various deserializers. * <p>Proper usage: read the field, probably in a hot loop. * <pre> * for( int row=0; row &lt; chunk._len; row++ ) * ...chunk.atd(row)... * </pre> **/ public transient int _len; /** Internal set of _len. Used by lots of subclasses. Not a publically visible API. */ int set_len(int len) { return _len = len; } /** Read-only length of chunk (number of rows). */ public int len() { return _len; } /** Normally==null, changed if chunk is written to. Not a publically readable or writable field. */ private transient Chunk _chk2; /** Exposed for internal testing only. Not a publically visible API. */ public Chunk chk2() { return _chk2; } /** Owning Vec; a read-only field */ transient Vec _vec; /** Owning Vec */ public Vec vec() { return _vec; } /** Set the owning Vec */ public void setVec(Vec vec) { _vec = vec; } /** Set the start */ public void setStart(long start) { _start = start; } /** The Big Data. Frequently set in the subclasses, but not otherwise a publically writable field. */ byte[] _mem; /** Short-cut to the embedded big-data memory. Generally not useful for * public consumption, since the data remains compressed and holding on to a * pointer to this array defeats the user-mode spill-to-disk. */ public byte[] getBytes() { return _mem; } public void setBytes(byte[] mem) { _mem = mem; } final long at8_abs(long i) { long x = i - (_start>0 ? _start : 0); if( 0 <= x && x < _len) return at8((int) x); throw new ArrayIndexOutOfBoundsException(""+_start+" <= "+i+" < "+(_start+ _len)); } /** Load a {@code double} value using absolute row numbers. Returns * Double.NaN if value is missing. * * <p>This version uses absolute element numbers, but must convert them to * chunk-relative indices - requiring a load from an aliasing local var, * leading to lower quality JIT'd code (similar issue to using iterator * objects). * * <p>Slightly slower than {@link #atd} since it range-checks within a chunk. * @return double value at the given row, or NaN if the value is missing */ final double at_abs(long i) { long x = i - (_start>0 ? _start : 0); if( 0 <= x && x < _len) return atd((int) x); throw new ArrayIndexOutOfBoundsException(""+_start+" <= "+i+" < "+(_start+ _len)); } /** Missing value status. * * <p>This version uses absolute element numbers, but must convert them to * chunk-relative indices - requiring a load from an aliasing local var, * leading to lower quality JIT'd code (similar issue to using iterator * objects). * * <p>Slightly slower than {@link #isNA} since it range-checks within a chunk. * @return true if the value is missing */ final boolean isNA_abs(long i) { long x = i - (_start>0 ? _start : 0); if( 0 <= x && x < _len) return isNA((int) x); throw new ArrayIndexOutOfBoundsException(""+_start+" <= "+i+" < "+(_start+ _len)); } /** Low half of a 128-bit UUID, or throws if the value is missing. * * <p>This version uses absolute element numbers, but must convert them to * chunk-relative indices - requiring a load from an aliasing local var, * leading to lower quality JIT'd code (similar issue to using iterator * objects). * * <p>Slightly slower than {@link #at16l} since it range-checks within a chunk. * @return Low half of a 128-bit UUID, or throws if the value is missing. */ final long at16l_abs(long i) { long x = i - (_start>0 ? _start : 0); if( 0 <= x && x < _len) return at16l((int) x); throw new ArrayIndexOutOfBoundsException(""+_start+" <= "+i+" < "+(_start+ _len)); } /** High half of a 128-bit UUID, or throws if the value is missing. * * <p>This version uses absolute element numbers, but must convert them to * chunk-relative indices - requiring a load from an aliasing local var, * leading to lower quality JIT'd code (similar issue to using iterator * objects). * * <p>Slightly slower than {@link #at16h} since it range-checks within a chunk. * @return High half of a 128-bit UUID, or throws if the value is missing. */ final long at16h_abs(long i) { long x = i - (_start>0 ? _start : 0); if( 0 <= x && x < _len) return at16h((int) x); throw new ArrayIndexOutOfBoundsException(""+_start+" <= "+i+" < "+(_start+ _len)); } /** String value using absolute row numbers, or null if missing. * * <p>This version uses absolute element numbers, but must convert them to * chunk-relative indices - requiring a load from an aliasing local var, * leading to lower quality JIT'd code (similar issue to using iterator * objects). * * <p>Slightly slower than {@link #atStr} since it range-checks within a chunk. * @return String value using absolute row numbers, or null if missing. */ final BufferedString atStr_abs(BufferedString bStr, long i) { long x = i - (_start>0 ? _start : 0); if( 0 <= x && x < _len) return atStr(bStr, (int) x); throw new ArrayIndexOutOfBoundsException(""+_start+" <= "+i+" < "+(_start+ _len)); } /** Load a {@code double} value using chunk-relative row numbers. Returns Double.NaN * if value is missing. * @return double value at the given row, or NaN if the value is missing */ public final double atd(int i) { return _chk2 == null ? atd_impl(i) : _chk2. atd_impl(i); } /** Load a {@code long} value using chunk-relative row numbers. Floating * point values are silently rounded to a long. Throws if the value is * missing. * @return long value at the given row, or throw if the value is missing */ public final long at8(int i) { return _chk2 == null ? at8_impl(i) : _chk2. at8_impl(i); } /** Missing value status using chunk-relative row numbers. * * @return true if the value is missing */ public final boolean isNA(int i) { return _chk2 == null ?isNA_impl(i) : _chk2.isNA_impl(i); } /** Low half of a 128-bit UUID, or throws if the value is missing. * * @return Low half of a 128-bit UUID, or throws if the value is missing. */ public final long at16l(int i) { return _chk2 == null ? at16l_impl(i) : _chk2.at16l_impl(i); } /** High half of a 128-bit UUID, or throws if the value is missing. * * @return High half of a 128-bit UUID, or throws if the value is missing. */ public final long at16h(int i) { return _chk2 == null ? at16h_impl(i) : _chk2.at16h_impl(i); } /** String value using chunk-relative row numbers, or null if missing. * * @return String value or null if missing. */ public final BufferedString atStr(BufferedString bStr, int i) { return _chk2 == null ? atStr_impl(bStr, i) : _chk2.atStr_impl(bStr, i); } public String stringAt(int i) { return atStr(new BufferedString(), i).toString(); } /** Write a {@code long} using absolute row numbers. There is no way to * write a missing value with this call. Under rare circumstances this can * throw: if the long does not fit in a double (value is larger magnitude * than 2^52), AND float values are stored in Vector. In this case, there * is no common compatible data representation. * * <p>As with all the {@code set} calls, if the value written does not fit * in the current compression scheme, the Chunk will be inflated into a * NewChunk and the value written there. Later, the NewChunk will be * compressed (after a {@link #close} call) and written back to the DKV. * i.e., there is some interesting cost if Chunk compression-types need to * change. * * <p>This version uses absolute element numbers, but must convert them to * chunk-relative indices - requiring a load from an aliasing local var, * leading to lower quality JIT'd code (similar issue to using iterator * objects). */ final void set_abs(long i, long l) { long x = i-_start; if (0 <= x && x < _len) set((int) x, l); else _vec.set(i,l); } /** Write a {@code double} using absolute row numbers; NaN will be treated as * a missing value. * * <p>As with all the {@code set} calls, if the value written does not fit * in the current compression scheme, the Chunk will be inflated into a * NewChunk and the value written there. Later, the NewChunk will be * compressed (after a {@link #close} call) and written back to the DKV. * i.e., there is some interesting cost if Chunk compression-types need to * change. * * <p>This version uses absolute element numbers, but must convert them to * chunk-relative indices - requiring a load from an aliasing local var, * leading to lower quality JIT'd code (similar issue to using iterator * objects). */ final void set_abs(long i, double d) { long x = i-_start; if (0 <= x && x < _len) set((int) x, d); else _vec.set(i,d); } /** Write a {@code float} using absolute row numbers; NaN will be treated as * a missing value. * * <p>As with all the {@code set} calls, if the value written does not fit * in the current compression scheme, the Chunk will be inflated into a * NewChunk and the value written there. Later, the NewChunk will be * compressed (after a {@link #close} call) and written back to the DKV. * i.e., there is some interesting cost if Chunk compression-types need to * change. * * <p>This version uses absolute element numbers, but must convert them to * chunk-relative indices - requiring a load from an aliasing local var, * leading to lower quality JIT'd code (similar issue to using iterator * objects). */ final void set_abs( long i, float f) { long x = i-_start; if (0 <= x && x < _len) set((int) x, f); else _vec.set(i,f); } /** Set the element as missing, using absolute row numbers. * * <p>As with all the {@code set} calls, if the value written does not fit * in the current compression scheme, the Chunk will be inflated into a * NewChunk and the value written there. Later, the NewChunk will be * compressed (after a {@link #close} call) and written back to the DKV. * i.e., there is some interesting cost if Chunk compression-types need to * change. * * <p>This version uses absolute element numbers, but must convert them to * chunk-relative indices - requiring a load from an aliasing local var, * leading to lower quality JIT'd code (similar issue to using iterator * objects). */ final void setNA_abs(long i) { long x = i-_start; if (0 <= x && x < _len) setNA((int) x); else _vec.setNA(i); } /** Set a {@code String}, using absolute row numbers. * * <p>As with all the {@code set} calls, if the value written does not fit * in the current compression scheme, the Chunk will be inflated into a * NewChunk and the value written there. Later, the NewChunk will be * compressed (after a {@link #close} call) and written back to the DKV. * i.e., there is some interesting cost if Chunk compression-types need to * change. * * <p>This version uses absolute element numbers, but must convert them to * chunk-relative indices - requiring a load from an aliasing local var, * leading to lower quality JIT'd code (similar issue to using iterator * objects). */ public final void set_abs(long i, String str) { long x = i-_start; if (0 <= x && x < _len) set((int) x, str); else _vec.set(i,str); } public final void set_abs(long i, UUID uuid) { long x = i-_start; if (0 <= x && x < _len) set((int) x, uuid); else _vec.set(i,uuid); } public boolean hasFloat(){return true;} public boolean hasNA(){return true;} /** Replace all rows with this new chunk */ public void replaceAll( Chunk replacement ) { assert _len == replacement._len; _vec.preWriting(); // One-shot writing-init _chk2 = replacement; assert _chk2._chk2 == null; // Replacement has NOT been written into } public Chunk deepCopy() { Chunk c2 = clone(); c2._vec=null; c2._start=-1; c2._cidx=-1; c2._mem = _mem.clone(); c2.initFromBytes(); assert len() == c2._len; return c2; } private void setWrite() { if( _chk2 != null ) return; // Already setWrite assert !(this instanceof NewChunk) : "Cannot direct-write into a NewChunk, only append"; setWrite(clone()); } private void setWrite(Chunk ck) { assert(_chk2==null); _vec.preWriting(); // One-shot writing-init _chk2 = ck; assert _chk2._chk2 == null; // Clone has NOT been written into } /** Write a {@code long} with check-relative indexing. There is no way to * write a missing value with this call. Under rare circumstances this can * throw: if the long does not fit in a double (value is larger magnitude * than 2^52), AND float values are stored in Vector. In this case, there * is no common compatible data representation. * * <p>As with all the {@code set} calls, if the value written does not fit * in the current compression scheme, the Chunk will be inflated into a * NewChunk and the value written there. Later, the NewChunk will be * compressed (after a {@link #close} call) and written back to the DKV. * i.e., there is some interesting cost if Chunk compression-types need to * change. * @return the set value */ public final long set(int idx, long l) { setWrite(); if( _chk2.set_impl(idx,l) ) return l; (_chk2 = inflate()).set_impl(idx,l); return l; } public final double [] set(double [] d){ assert d.length == _len && _chk2 == null; setWrite(new NewChunk(this,d)); return d; } /** Write a {@code double} with check-relative indexing. NaN will be treated * as a missing value. * * <p>As with all the {@code set} calls, if the value written does not fit * in the current compression scheme, the Chunk will be inflated into a * NewChunk and the value written there. Later, the NewChunk will be * compressed (after a {@link #close} call) and written back to the DKV. * i.e., there is some interesting cost if Chunk compression-types need to * change. * @return the set value */ public final double set(int idx, double d) { setWrite(); if( _chk2.set_impl(idx,d) ) return d; (_chk2 = inflate()).set_impl(idx,d); return d; } /** Write a {@code float} with check-relative indexing. NaN will be treated * as a missing value. * * <p>As with all the {@code set} calls, if the value written does not fit * in the current compression scheme, the Chunk will be inflated into a * NewChunk and the value written there. Later, the NewChunk will be * compressed (after a {@link #close} call) and written back to the DKV. * i.e., there is some interesting cost if Chunk compression-types need to * change. * @return the set value */ public final float set(int idx, float f) { setWrite(); if( _chk2.set_impl(idx,f) ) return f; (_chk2 = inflate()).set_impl(idx,f); return f; } /** Set a value as missing. * * <p>As with all the {@code set} calls, if the value written does not fit * in the current compression scheme, the Chunk will be inflated into a * NewChunk and the value written there. Later, the NewChunk will be * compressed (after a {@link #close} call) and written back to the DKV. * i.e., there is some interesting cost if Chunk compression-types need to * change. * @return the set value */ public final boolean setNA(int idx) { setWrite(); if( _chk2.setNA_impl(idx) ) return true; (_chk2 = inflate()).setNA_impl(idx); return true; } /** Write a {@code String} with check-relative indexing. {@code null} will * be treated as a missing value. * * <p>As with all the {@code set} calls, if the value written does not fit * in the current compression scheme, the Chunk will be inflated into a * NewChunk and the value written there. Later, the NewChunk will be * compressed (after a {@link #close} call) and written back to the DKV. * i.e., there is some interesting cost if Chunk compression-types need to * change. * @return the set value */ public final String set(int idx, String str) { setWrite(); if( _chk2.set_impl(idx,str) ) return str; (_chk2 = inflate()).set_impl(idx,str); return str; } public final UUID set(int idx, UUID uuid) { setWrite(); long lo = uuid.getLeastSignificantBits(); long hi = uuid.getMostSignificantBits(); if( _chk2.set_impl(idx, lo, hi) ) return uuid; _chk2 = inflate(); _chk2.set_impl(idx,lo, hi); return uuid; } private Object setUnknown(int idx) { setNA(idx); return null; } /** * @param idx index of the value in Chunk * @param x new value to set * @return x on success, or null if something went wrong */ public final Object setAny(int idx, Object x) { return x instanceof String ? set(idx, (String) x) : x instanceof Double ? set(idx, (Double)x) : x instanceof Float ? set(idx, (Float)x) : x instanceof Long ? set(idx, (Long)x) : x instanceof Integer ? set(idx, ((Integer)x).longValue()) : x instanceof UUID ? set(idx, (UUID) x) : x instanceof java.util.Date ? set(idx, ((java.util.Date) x).getTime()) : /* otherwise */ setUnknown(idx); } /** After writing we must call close() to register the bulk changes. If a * NewChunk was needed, it will be compressed into some other kind of Chunk. * The resulting Chunk (either a modified self, or a compressed NewChunk) * will be written to the DKV. Only after that {@code DKV.put} completes * will all readers of this Chunk witness the changes. * @return the passed-in {@link Futures}, for flow-coding. */ public Futures close( int cidx, Futures fs ) { if( this instanceof NewChunk ) _chk2 = this; if( _chk2 == null ) return fs; // No change? if( _chk2 instanceof NewChunk ) _chk2 = ((NewChunk)_chk2).new_close(); DKV.put(_vec.chunkKey(cidx),_chk2,fs,true); // Write updated chunk back into K/V return fs; } /** @return Chunk index */ public int cidx() { assert _cidx != -1 : "Chunk idx was not properly loaded!"; return _cidx; } public final Chunk setVolatile(double [] ds) { Chunk res; Value v = new Value(_vec.chunkKey(_cidx), res = new C8DVolatileChunk(ds),ds.length*8,Value.ICE); DKV.put(v._key,v); return res; } public final Chunk setVolatile(int[] vals) { Chunk res; Value v = new Value(_vec.chunkKey(_cidx), res = new C4VolatileChunk(vals),vals.length*4,Value.ICE); DKV.put(v._key,v); return res; } public boolean isVolatile() {return false;} static class WrongType extends IllegalArgumentException { private final Class<?> expected; private final Class<?> actual; public WrongType(Class<?> expected, Class<?> actual) { super("Expected: " + expected + ", actual: " + actual); this.expected = expected; this.actual = actual; } } static WrongType wrongType(Class<?> expected, Class<?> actual) { return new WrongType(expected, actual); } /** Chunk-specific readers. Not a public API */ abstract double atd_impl(int idx); abstract long at8_impl(int idx); abstract boolean isNA_impl(int idx); long at16l_impl(int idx) { throw wrongType(UUID.class, Object.class); } long at16h_impl(int idx) { throw wrongType(UUID.class, Object.class); } BufferedString atStr_impl(BufferedString bStr, int idx) { throw new IllegalArgumentException("Not a String"); } /** Chunk-specific writer. Returns false if the value does not fit in the * current compression scheme. */ abstract boolean set_impl (int idx, long l ); abstract boolean set_impl (int idx, double d ); abstract boolean set_impl (int idx, float f ); abstract boolean setNA_impl(int idx); boolean set_impl (int idx, String str) { return false; } boolean set_impl(int i, long lo, long hi) { return false; } //Zero sparse methods: /** Sparse Chunks have a significant number of zeros, and support for * skipping over large runs of zeros in a row. * @return true if this Chunk is sparse. */ public boolean isSparseZero() {return false;} /** Sparse Chunks have a significant number of zeros, and support for * skipping over large runs of zeros in a row. * @return At least as large as the count of non-zeros, but may be significantly smaller than the {@link #_len} */ public int sparseLenZero() {return _len;} /** * Skips a section of either NAs or Zeros in a sparse chunk. * Note: This method can only be used when NAs and Zeros have the same meaning to the caller! * @param rid Start search from this index (excluded). * @return Index of a next non-sparse value (this can include NA and Zero * depending on if the chunk is zero-sparse or na-sparse). */ public int nextNZ(int rid){ return rid + 1;} /** * Version of nextNZ() that allows caller to prevent skipping NAs. * @param rid Start search from this index (excluded). * @param onlyTrueZero if true, only actual Zeros can be skipped. NA-sparse chunks will be treated as dense. * @return Index of a next non-sparse value. */ public int nextNZ(int rid, boolean onlyTrueZero) { return rid + 1; } /** * Get indeces of non-zero values stored in this chunk * @return array of chunk-relative indices of values stored in this chunk. */ public int nonzeros(int [] res) { int k = 0; for( int i = 0; i < _len; ++i) if(atd(i) != 0) res[k++] = i; return k; } //NA sparse methods: /** Sparse Chunks have a significant number of NAs, and support for * skipping over large runs of NAs in a row. * @return true if this Chunk is sparseNA. */ public boolean isSparseNA() {return false;} /** Sparse Chunks have a significant number of NAs, and support for * skipping over large runs of NAs in a row. * @return At least as large as the count of non-NAs, but may be significantly smaller than the {@link #_len} */ public int sparseLenNA() {return _len;} /** Report the Chunk min-value (excluding NAs), or NaN if unknown. Actual * min can be higher than reported. Used to short-cut RollupStats for * constant and boolean chunks. */ double min() { return Double.NaN; } /** Report the Chunk max-value (excluding NAs), or NaN if unknown. Actual * max can be lower than reported. Used to short-cut RollupStats for * constant and boolean chunks. */ double max() { return Double.NaN; } public final NewChunk inflate(){ return extractRows(new NewChunk(this), 0,_len);} /** Return the next Chunk, or null if at end. Mostly useful for parsers or * optimized stencil calculations that want to "roll off the end" of a * Chunk, but in a highly optimized way. */ public Chunk nextChunk( ) { return _vec.nextChunk(this); } /** @return String version of a Chunk, class name and range*/ @Override public String toString() { return getClass().getSimpleName() + "[" + _start + ".." + (_start + _len - 1) + "]"; } /** In memory size in bytes of the compressed Chunk plus embedded array. */ public long byteSize() { long s= _mem == null ? 0 : _mem.length; s += (2+5)*8 + 12; // 2 hdr words, 5 other words, @8bytes each, plus mem array hdr if( _chk2 != null ) s += _chk2.byteSize(); return s; } /** Custom serializers implemented by Chunk subclasses: the _mem field * contains ALL the fields already. */ public final AutoBuffer write_impl(AutoBuffer bb) {return bb.putA1(_mem);} @Override public byte [] asBytes(){return _mem;} @Override public final Chunk reloadFromBytes(byte [] ary){ _mem = ary; initFromBytes(); return this; } protected abstract void initFromBytes(); public final Chunk read_impl(AutoBuffer ab){ _mem = ab.getA1(); initFromBytes(); return this; } // /** Custom deserializers, implemented by Chunk subclasses: the _mem field // * contains ALL the fields already. Init _start to -1, so we know we have // * not filled in other fields. Leave _vec and _chk2 null, leave _len // * unknown. */ // abstract public Chunk read_impl( AutoBuffer ab ); // ----------------- // Support for fixed-width format printing // private String pformat () { return pformat0(); } // private int pformat__len { return pformat_len0(); } /** Fixed-width format printing support. Filled in by the subclasses. */ public byte precision() { return -1; } // Digits after the decimal, or -1 for "all" // protected String pformat0() { // long min = (long)_vec.min(); // if( min < 0 ) return "% "+pformat_len0()+"d"; // return "%"+pformat_len0()+"d"; // } // protected int pformat_len0() { // int len=0; // long min = (long)_vec.min(); // if( min < 0 ) len++; // long max = Math.max(Math.abs(min),Math.abs((long)_vec.max())); // throw H2O.unimpl(); // //for( int i=1; i<DParseTask.powers10i.length; i++ ) // // if( max < DParseTask.powers10i[i] ) // // return i+len; // //return 20; // } // protected int pformat_len0( double scale, int lg ) { // double dx = Math.log10(scale); // int x = (int)dx; // throw H2O.unimpl(); // //if( DParseTask.pow10i(x) != scale ) throw H2O.unimpl(); // //int w=1/*blank/sign*/+lg/*compression limits digits*/+1/*dot*/+1/*e*/+1/*neg exp*/+2/*digits of exp*/; // //return w; // } /** Used by the parser to help report various internal bugs. Not intended for public use. */ public final void reportBrokenCategorical(int i, int j, long l, int[] cmap, int levels) { StringBuilder sb = new StringBuilder("Categorical renumber task, column # " + i + ": Found OOB index " + l + " (expected 0 - " + cmap.length + ", global domain has " + levels + " levels) pulled from " + getClass().getSimpleName() + "\n"); int k = 0; for(; k < Math.min(5,_len); ++k) sb.append("at8_abs[" + (k+_start) + "] = " + atd(k) + ", _chk2 = " + (_chk2 != null?_chk2.atd(k):"") + "\n"); k = Math.max(k,j-2); sb.append("...\n"); for(; k < Math.min(_len,j+2); ++k) sb.append("at8_abs[" + (k+_start) + "] = " + atd(k) + ", _chk2 = " + (_chk2 != null?_chk2.atd(k):"") + "\n"); sb.append("...\n"); k = Math.max(k,_len-5); for(; k < _len; ++k) sb.append("at8_abs[" + (k+_start) + "] = " + atd(k) + ", _chk2 = " + (_chk2 != null?_chk2.atd(k):"") + "\n"); throw new RuntimeException(sb.toString()); } public abstract <T extends ChunkVisitor> T processRows(T v, int from, int to); public abstract <T extends ChunkVisitor> T processRows(T v, int [] ids); // convenience methods wrapping around visitor interface public NewChunk extractRows(NewChunk nc, int from, int to){ return processRows(new ChunkVisitor.NewChunkVisitor(nc),from,to)._nc; } public NewChunk extractRows(NewChunk nc, int[] rows){ return processRows(new ChunkVisitor.NewChunkVisitor(nc),rows)._nc; } public NewChunk extractRows(NewChunk nc, int row){ return processRows(new ChunkVisitor.NewChunkVisitor(nc),row,row+1)._nc; } /** * Dense bulk interface, fetch values from the given range * @param vals * @param from * @param to */ public double [] getDoubles(double[] vals, int from, int to){ return getDoubles(vals,from,to, Double.NaN);} public double [] getDoubles(double [] vals, int from, int to, double NA){ return processRows(new ChunkVisitor.DoubleAryVisitor(vals,NA),from,to).vals; } public double[] getDoubles() { return getDoubles(MemoryManager.malloc8d(_len), 0, _len); } public int [] getIntegers(int [] vals, int from, int to, int NA){ return processRows(new ChunkVisitor.IntAryVisitor(vals,NA),from,to).vals; } /** * Dense bulk interface, fetch values from the given ids * @param vals * @param ids */ public double[] getDoubles(double [] vals, int [] ids){ return processRows(new ChunkVisitor.DoubleAryVisitor(vals),ids).vals; } /** * Sparse bulk interface, stream through the compressed values and extract them into dense double array. * @param vals holds extracted values, length must be >= this.sparseLen() * @param ids holds extracted chunk-relative row ids, length must be >= this.sparseLen() * @return number of extracted (non-zero) elements, equal to sparseLen() */ public int getSparseDoubles(double[] vals, int[] ids){return getSparseDoubles(vals,ids,Double.NaN);} public int getSparseDoubles(double [] vals, int [] ids, double NA) { return processRows(new ChunkVisitor.SparseDoubleAryVisitor(vals,ids,isSparseNA(),NA),0,_len).sparseLen(); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/ChunkUtils.java
package water.fvec; import water.DKV; import water.Key; /** * Simple helper class which publishes some frame and chunk package private methods as public */ public class ChunkUtils { public static NewChunk[] createNewChunks(String name, byte[] vecTypes, int chunkId){ return Frame.createNewChunks(name, vecTypes, chunkId); } public static NewChunk[] createNewChunks(String name, byte[] vecTypes, int chunkId, boolean[] sparse) { return Frame.createNewChunks(name, vecTypes, chunkId, sparse); } public static void closeNewChunks(NewChunk[] nchks){ Frame.closeNewChunks(nchks); } public static Chunk[] getChunks(Frame fr, int cidx) { Chunk[] chunks = new Chunk[fr.vecs().length]; for(int i=0; i<fr.vecs().length; i++){ chunks[i] = fr.vec(i).chunkForChunkIdx(cidx); } return chunks; } public static void initFrame(String keyName, String[] names) { Frame fr = new water.fvec.Frame(Key.<Frame>make(keyName)); fr.preparePartialFrame(names); // Save it directly to DKV fr.update(); } public static Frame finalizeFrame(String keyName, long[] rowsPerChunk, byte[] colTypes, String[][] colDomains) { return finalizeFrame(keyName, rowsPerChunk, colTypes, colDomains, true); } public static Frame finalizeFrame(String keyName, long[] rowsPerChunk, byte[] colTypes, String[][] colDomains, boolean unlock) { Frame fr = DKV.getGet(keyName); fr.finalizePartialFrame(rowsPerChunk, colDomains, colTypes, unlock); return fr; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/ChunkVisitor.java
package water.fvec; import water.parser.BufferedString; import water.util.PrettyPrint; /** * Created by tomas on 3/8/17. * Base class for using visitor pattern with chunks. */ public abstract class ChunkVisitor { public boolean expandedVals() { return false; } void addValue(BufferedString bs) { throw new UnsupportedOperationException(); } void addValue(long uuid_lo, long uuid_hi) { throw new UnsupportedOperationException(); } void addValue(int val) { throw new UnsupportedOperationException(); } void addValue(double val) { throw new UnsupportedOperationException(); } void addValue(long val) { throw new UnsupportedOperationException(); } void addValue(long m, int e) { addValue(PrettyPrint.pow10(m,e)); } void addZeros(int zeros) { throw new UnsupportedOperationException(); } void addNAs(int nas) { throw new UnsupportedOperationException(); } /** * Visitor wrapper around NewChunk. Usefull for extracting rows from chunks. */ public static final class NewChunkVisitor extends ChunkVisitor { final NewChunk _nc; public NewChunkVisitor(NewChunk nc){_nc = nc;} @Override public boolean expandedVals(){return true;} @Override void addValue(BufferedString bs){_nc.addStr(bs);} @Override void addValue(long uuid_lo, long uuid_hi){_nc.addUUID(uuid_lo,uuid_hi);} @Override void addValue(int val) {_nc.addNum(val,0);} @Override void addValue(long val) {_nc.addNum(val,0);} @Override void addValue(long val, int exp) {_nc.addNum(val,exp);} @Override void addValue(double val) {_nc.addNum(val);} @Override void addZeros(int zeros) {_nc.addZeros(zeros);} @Override void addNAs(int nas) {_nc.addNAs(nas);} } /** * Simple chunk visitor for extracting rows from chunks into a double array. */ public static final class DoubleAryVisitor extends ChunkVisitor { public final double [] vals; private int _k = 0; private final double _na; DoubleAryVisitor(double [] vals){this(vals,Double.NaN);} DoubleAryVisitor(double [] vals, double NA){ this.vals = vals; _na = NA;} @Override void addValue(int val) { vals[_k++] = val;} @Override void addValue(long val) { vals[_k++] = val;} @Override void addValue(double val) { vals[_k++] = Double.isNaN(val)?_na:val;} @Override void addZeros(int zeros) { int k = _k; int kmax = k +zeros; for(;k < kmax; k++) vals[k] = 0; _k = kmax; } @Override void addNAs(int nas) { int k = _k; int kmax = k + nas; for(;k < kmax; k++) vals[k] = _na; _k = kmax; } } /** * Simple chunk visitor for extracting rows from chunks into a sparse double array. */ public static final class SparseDoubleAryVisitor extends ChunkVisitor { public final boolean naSparse; public final double [] vals; public final int [] ids; private int _sparseLen; private int _len; private final double _na; public int sparseLen(){return _sparseLen;} SparseDoubleAryVisitor(double [] vals, int [] ids){this(vals,ids,false,Double.NaN);} SparseDoubleAryVisitor(double [] vals, int [] ids, boolean naSparse){this(vals, ids, naSparse, Double.NaN);} SparseDoubleAryVisitor(double [] vals, int [] ids, boolean naSparse, double NA){this.vals = vals; this.ids = ids; _na = NA; this.naSparse = naSparse;} @Override void addValue(int val) {ids[_sparseLen] = _len++; vals[_sparseLen++] = val;} @Override void addValue(long val) {ids[_sparseLen] = _len++; vals[_sparseLen++] = val;} @Override void addValue(double val) {ids[_sparseLen] = _len++; vals[_sparseLen++] = Double.isNaN(val)?_na:val;} @Override void addZeros(int zeros) { if(naSparse) { int kmax = _sparseLen + zeros; for (int k = _sparseLen; k < kmax; k++) { ids[k] = _len++; vals[k] = 0; } _sparseLen = kmax; } else _len += zeros; } @Override void addNAs(int nas) { if(!naSparse) { int kmax = _sparseLen + nas; for (int k = _sparseLen; k < kmax; k++) { ids[k] = _len++; vals[k] = _na; } _sparseLen = kmax; } else _len += nas; } } /** * Chunk visitor for combining values from chunk with values from a given double array */ public static final class CombiningDoubleAryVisitor extends ChunkVisitor { public final double [] vals; private int _k = 0; private final double _na; public CombiningDoubleAryVisitor(double [] vals){this(vals,Double.NaN);} CombiningDoubleAryVisitor(double [] vals, double NA){ this.vals = vals; _na = NA;} @Override void addValue(int val) { vals[_k++] += val;} @Override void addValue(long val) { vals[_k++] += val;} @Override void addValue(double val) { if (Double.isNaN(val)) vals[_k++] = _na; else vals[_k++] += val;} @Override void addZeros(int zeros) { _k += zeros; } @Override void addNAs(int nas) { int k = _k; int kmax = k + nas; for(;k < kmax; k++) vals[k] = _na; _k = kmax; } public void reset() { _k = 0; } } /** * Simple chunk visitor for extracting rows from chunks into a integer array. */ public static final class IntAryVisitor extends ChunkVisitor { public final int [] vals; private int _k = 0; private final int _na; IntAryVisitor(int [] vals, int NA){this.vals = vals; _na = NA;} @Override public void addValue(int val) {vals[_k++] = val;} @Override public void addValue(long val) { if(Integer.MAX_VALUE < val || val < Integer.MIN_VALUE) throw new RuntimeException(val + " does not fit into int"); vals[_k++] = (int)val; } @Override public void addValue(double val) { if (Double.isNaN(val)) { vals[_k++] = _na; } else { int i = (int) val; if (i != val) throw new RuntimeException(val + " does not fit into int"); vals[_k++] = i; } } @Override public void addZeros(int zeros) { int k = _k; int kmax = k +zeros; for(;k < kmax; k++)vals[k] = 0; _k = kmax; } @Override public void addNAs(int nas) { int k = _k; int kmax = k + nas; for(;k < kmax; k++)vals[k] = _na; _k = kmax; } } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/CreateInteractions.java
package water.fvec; import hex.Interaction; import jsr166y.CountedCompleter; import water.*; import water.util.IcedHashMap; import water.util.IcedLong; import java.util.*; /** * Helper to create interaction features between categorical columns */ public class CreateInteractions extends H2O.H2OCountedCompleter { public CreateInteractions(Interaction ci) { _job = ci._job._key; _ci = ci; } // used for testing public CreateInteractions(int maxFactors, int minOccurrence) { _job=null; _ci=new Interaction(); _ci._max_factors=maxFactors; _ci._min_occurrence=minOccurrence; } final private Interaction _ci; static final private int _missing = Integer.MIN_VALUE; //marker for missing factor level static final String _other = "other"; // name for lost factor levels private Frame _target; final private Key<Job> _job; private Map<Long, Long> _sortedMap = null; private static Map<Long, Long> mySort(Map<IcedLong, IcedLong> unsortMap) { List<Map.Entry<IcedLong, IcedLong>> list = new LinkedList<>(unsortMap.entrySet()); // Sorting the list based on values Collections.sort(list, new Comparator<Map.Entry<IcedLong, IcedLong>>() { public int compare(Map.Entry<IcedLong, IcedLong> o1, Map.Entry<IcedLong, IcedLong> o2) { return ((Long)o2.getValue()._val).compareTo(o1.getValue()._val); } }); // Maintaining insertion order with the help of LinkedList Map sortedMap = new LinkedHashMap<>(); for (Map.Entry<IcedLong, IcedLong> entry : list) { sortedMap.put(entry.getKey()._val, entry.getValue()._val); } return sortedMap; } // Create a combined domain from the categorical values that map to domain A and domain B // Both categorical integers are combined into a long = (int,int), and the unsortedMap keeps the occurrence count for each pair-wise interaction public String[] makeDomain(Map<IcedLong, IcedLong> unsortedMap, String[] dA, String[] dB) { String[] _domain; // Log.info("Collected hash table"); // Log.info(java.util.Arrays.deepToString(unsortedMap.entrySet().toArray())); // Log.info("Interaction between " + dA.length + " and " + dB.length + " factor levels => " + // ((long)dA.length * dB.length) + " possible factors."); _sortedMap = mySort(unsortedMap); // create domain of the most frequent unique factors long factorCount = 0; // Log.info("Found " + _sortedMap.size() + " unique interaction factors (out of " + ((long)dA.length * (long)dB.length) + ")."); _domain = new String[_sortedMap.size()]; //TODO: use ArrayList here, then convert to array Iterator it2 = _sortedMap.entrySet().iterator(); int d = 0; while (it2.hasNext()) { Map.Entry kv = (Map.Entry)it2.next(); final long ab = (Long)kv.getKey(); final long count = (Long)kv.getValue(); if (factorCount < _ci._max_factors && count >= _ci._min_occurrence) { factorCount++; // extract the two original factor categoricals String feature = ""; if (dA != dB) { int a = (int)(ab >> 32); final String fA = a != _missing ? dA[a] : "NA"; feature = fA + "_"; } int b = (int) ab; String fB = b != _missing ? dB[b] : "NA"; feature += fB; // Log.info("Adding interaction feature " + feature + ", occurrence count: " + count); // Log.info("Total number of interaction factors so far: " + factorCount); _domain[d++] = feature; } else break; } if (d < _sortedMap.size()) { // Log.info("Truncated map to " + _sortedMap.size() + " elements."); String[] copy = new String[d+1]; System.arraycopy(_domain, 0, copy, 0, d); copy[d] = _other; _domain = copy; Map tm = new LinkedHashMap<>(); it2 = _sortedMap.entrySet().iterator(); while (--d >= 0) { Map.Entry kv = (Map.Entry) it2.next(); tm.put(kv.getKey(), kv.getValue()); } _sortedMap = tm; } // Log.info("Created domain: " + Arrays.deepToString(_domain)); return _domain; } private ArrayList<int[]> interactions() { ArrayList<int[]> al = new ArrayList<>(); if (!_ci._pairwise || _ci._factors.length < 3) { al.add(_ci._factors); } else { // pair-wise for (int i = 0; i < _ci._factors.length; ++i) { for (int j = i + 1; j < _ci._factors.length; ++j) { al.add(new int[]{_ci._factors[i], _ci._factors[j]}); } } } return al; } public int work() { ArrayList<int[]> al = interactions(); int work=0; for (int l=0; l<al.size(); ++l) { int[] factors = al.get(l); int start = factors.length == 1 ? 0 : 1; for (int i = start; i < factors.length; ++i) { work++; } } return work; } @Override public void compute2() { DKV.remove(_ci._job._result); Frame source_frame = DKV.getGet(_ci._source_frame); ArrayList<int[]> al = interactions(); for (int l=0; l<al.size(); ++l) { int[] factors = al.get(l); int idx1 = factors[0]; Vec tmp = null; int start = factors.length == 1 ? 0 : 1; Frame _out = null; for (int i = start; i < factors.length; ++i) { String name; int idx2 = factors[i]; if (i > 1) { idx1 = _out.find(tmp); assert idx1 >= 0; name = _out._names[idx1] + "_" + source_frame._names[idx2]; } else { name = source_frame._names[idx1] + "_" + source_frame._names[idx2]; } // Log.info("Combining columns " + idx1 + " and " + idx2); final Vec A = i > 1 ? _out.vecs()[idx1] : source_frame.vecs()[idx1]; final Vec B = source_frame.vecs()[idx2]; // Pass 1: compute unique domains of all interaction features createInteractionDomain pass1 = new createInteractionDomain(A._key.equals(B._key), _ci._interactOnNA).doAll(A, B); // Create a new Vec based on the domain final Vec vec = source_frame.anyVec().makeZero(makeDomain(pass1._unsortedMap, A.domain(), B.domain())); if (i > 1) { _out.add(name, vec); } else { assert(_out == null); _out = new Frame(new String[]{name}, new Vec[]{vec}); } final Vec C = _out.lastVec(); // Create array of categorical pairs, in the same (sorted) order as in the _domain map -> for linear lookup // Note: "other" is not mapped in keys, so keys.length can be 1 less than domain.length long[] keys = new long[_sortedMap.size()]; int pos = 0; for (long k : _sortedMap.keySet()) { keys[pos++] = k; } assert (C.domain().length == keys.length || C.domain().length == keys.length + 1); // domain might contain _other // Pass 2: fill Vec values new fillInteractionCategoricals(A._key.equals(B._key), keys).doAll(A, B, C); tmp = C; // remove temporary vec if (i > 1) { final int idx = _out.vecs().length - 2; //second-last vec // Log.info("Removing column " + _out._names[idx]); _out.remove(idx).remove(); } _ci._job.update(1); } if (_target == null) { _target = new Frame(_ci._job._result, _out.names(), _out.vecs()); _target.delete_and_lock(_job); } else { _target.add(_out); } } tryComplete(); } @Override public void onCompletion(CountedCompleter caller) { _target.update(_job); _target.unlock(_job); } // Create interaction domain public static class createInteractionDomain extends MRTask<createInteractionDomain> { // INPUT final private boolean _same; // self interaction final private boolean _interactOnNA; // allow NAs to count as lvls final private int[] _restrictedEnumA; final private int[] _restrictedEnumB; // OUTPUT private IcedHashMap<IcedLong, IcedLong> _unsortedMap = null; public IcedHashMap<IcedLong, IcedLong> getMap() { return _unsortedMap; } public createInteractionDomain(boolean same, boolean interactOnNA) { _same = same; _interactOnNA=interactOnNA; _restrictedEnumA = _restrictedEnumB =null; } // TODO: continue to extend functionality here and bridge to InteractionWrappedVec so that code can be shared public createInteractionDomain(boolean same, boolean interactOnNA, int[] restrictedEnumLeft, int[] restrictedEnumRite) { _same = same; _interactOnNA=interactOnNA; _restrictedEnumA =restrictedEnumLeft; _restrictedEnumB =restrictedEnumRite; } @Override public void map(Chunk A, Chunk B) { _unsortedMap = new IcedHashMap<>(); // find unique interaction domain HashSet<Integer> restrictedA = _restrictedEnumA==null?null: new HashSet<Integer>(), restrictedB = _restrictedEnumB==null?null: new HashSet<Integer>(); if( restrictedA!=null ) for (int i: _restrictedEnumA) restrictedA.add(i); if( restrictedB!=null ) for (int i: _restrictedEnumB) restrictedB.add(i); for (int r = 0; r < A._len; r++) { int a = A.isNA(r) ? _missing : (int)A.at8(r); if( !_interactOnNA && a==_missing ) continue; // most readable way to express if( restrictedA!=null && !restrictedA.contains(a) ) continue; // not part of the limited set long ab; if (!_same) { int b = B.isNA(r) ? _missing : (int)B.at8(r); if( !_interactOnNA && b==_missing ) continue; if( restrictedB!=null && !restrictedB.contains(b) ) continue; // not part of the limited set // key: combine both ints into a long ab = ((long) a << 32) | (b & 0xFFFFFFFFL); assert a == (int) (ab >> 32); assert b == (int) ab; } else { if (a == _missing) continue; ab = (long)a; } // add key to hash map, and count occurrences (for pruning) IcedLong AB = new IcedLong(ab); if (_unsortedMap.containsKey(AB)) { _unsortedMap.get(AB)._val++; } else { _unsortedMap.put(AB, new IcedLong(1)); } } } @Override public void reduce(createInteractionDomain mrt) { assert(mrt._unsortedMap != null); assert(_unsortedMap != null); for (Map.Entry<IcedLong,IcedLong> e : mrt._unsortedMap.entrySet()) { IcedLong x = _unsortedMap.get(e.getKey()); if (x != null) { x._val+=e.getValue()._val; } else { _unsortedMap.put(e.getKey(), e.getValue()); } } mrt._unsortedMap = null; // Log.info("Merged hash tables"); // Log.info(java.util.Arrays.deepToString(_unsortedMap.entrySet().toArray())); } } // Fill interaction categoricals in last Vec in Frame private static class fillInteractionCategoricals extends MRTask<fillInteractionCategoricals> { // INPUT boolean _same; final long[] _keys; //minimum information to be sent over the wire transient private java.util.List<java.util.Map.Entry<Long,Integer>> _valToIndex; //node-local shared helper for binary search public fillInteractionCategoricals(boolean same, long[] keys) { _same = same; _keys = keys; } @Override protected void setupLocal() { // turn _keys into a sorted array of pairs _valToIndex = new java.util.ArrayList<>(); // map factor level (int,int) to domain index (long) for (int i=0;i<_keys.length;++i) { _valToIndex.add(new AbstractMap.SimpleEntry<>(_keys[i], i)); } // sort by key (the factor level) Collections.sort(_valToIndex, new Comparator<Map.Entry<Long, Integer>>() { @Override public int compare(Map.Entry<Long, Integer> o1, Map.Entry<Long, Integer> o2) { return o1.getKey().compareTo(o2.getKey()); } }); } @Override public void map(Chunk A, Chunk B, Chunk C) { // find unique interaction domain for (int r = 0; r < A._len; r++) { final int a = A.isNA(r) ? _missing : (int)A.at8(r); long ab; if (!_same) { final int b = B.isNA(r) ? _missing : (int) B.at8(r); ab = ((long) a << 32) | (b & 0xFFFFFFFFL); // key: combine both ints into a long } else { ab = (long)a; } if (_same && A.isNA(r)) { C.setNA(r); } else { // find _domain index for given factor level ab int level = -1; int pos = Collections.binarySearch(_valToIndex, new AbstractMap.SimpleEntry<Long,Integer>(ab,0), new Comparator<Map.Entry<Long, Integer>>() { @Override public int compare(Map.Entry<Long, Integer> o1, Map.Entry<Long, Integer> o2) { return o1.getKey().compareTo(o2.getKey()); } }); if (pos >= 0) { level = _valToIndex.get(pos).getValue(); assert _keys[level] == ab; //confirm that binary search in _valToIndex worked } if (level < 0) { for (int i=0; i<_keys.length; ++i) { assert (_keys[i] != ab); } level = _fr.lastVec().domain().length-1; assert _fr.lastVec().domain()[level].equals(_other); } C.set(r, level); } } } } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/FileVec.java
package water.fvec; import water.*; import water.util.Log; import water.util.MathUtils; import water.util.UnsafeUtils; public abstract class FileVec extends ByteVec { long _len; // File length private final byte _be; public byte getBackend() { return _be; } public String getPath() { return getPathForKey(_key); } // Returns String with path for given key. public static String getPathForKey(Key k) { final int off = k._kb[0]==Key.CHK || k._kb[0]==Key.VEC ? Vec.KEY_PREFIX_LEN : 0; String p = new String(k._kb,off,k._kb.length-off); if(p.startsWith("nfs:/")) p = p.substring("nfs:/".length()); else if (p.startsWith("nfs:\\")) p = p.substring("nfs:\\".length()); return p; } /** Log-2 of Chunk size. */ public static final int DFLT_LOG2_CHUNK_SIZE = 20/*1Meg*/+2/*4Meg*/; /** Default Chunk size in bytes, useful when breaking up large arrays into * "bite-sized" chunks. Bigger increases batch sizes, lowers overhead * costs, lower increases fine-grained parallelism. */ public static final int DFLT_CHUNK_SIZE = 1 << DFLT_LOG2_CHUNK_SIZE; public int _chunkSize = DFLT_CHUNK_SIZE; public int _nChunks = -1; protected FileVec(Key key, long len, byte be) { super(key,-1/*no rowLayout*/); _len = len; _be = be; } public void setNChunks(int n){ _nChunks = n; setChunkSize((int)length()/n); } /** * Chunk size must be positive, 1G or less, and a power of two. * Any values that aren't a power of two will be reduced to the * first power of two lower than the provided chunkSize. * <p> * Since, optimal chunk size is not known during FileVec instantiation, * setter is required to both set it, and keep it in sync with * _log2ChkSize. * </p> * @param chunkSize requested chunk size to be used when parsing * @return actual _chunkSize setting */ public int setChunkSize(int chunkSize) { return setChunkSize(null, chunkSize); } public int setChunkSize(Frame fr, int chunkSize) { // Clear cached chunks first // Peeking into a file before the chunkSize has been set // will load chunks of the file in DFLT_CHUNK_SIZE amounts. // If this side-effect is not reversed when _chunkSize differs // from the default value, parsing will either double read // sections (_chunkSize < DFLT_CHUNK_SIZE) or skip data // (_chunkSize > DFLT_CHUNK_SIZE). This reverses this side-effect. Futures fs = new Futures(); Keyed.remove(_key, fs, true); fs.blockForPending(); if (chunkSize <= 0) throw new IllegalArgumentException("Chunk sizes must be > 0."); if (chunkSize > (1<<30) ) throw new IllegalArgumentException("Chunk sizes must be < 1G."); _chunkSize = chunkSize; //Now reset the chunk size on each node fs = new Futures(); DKV.put(_key, this, fs); // also update Frame to invalidate local caches if (fr != null ) { fr.reloadVecs(); DKV.put(fr._key, fr, fs); } fs.blockForPending(); return _chunkSize; } @Override public long length() { return _len; } @Override public int nChunks() { if(_nChunks != -1) // number of chunks can be set explicitly return _nChunks; return (int)Math.max(1,_len / _chunkSize + ((_len % _chunkSize != 0)?1:0)); } @Override public int nonEmptyChunks() { return nChunks(); } @Override public boolean writable() { return false; } /** Size of vector data. */ @Override public long byteSize(){return length(); } // Convert a row# to a chunk#. For constant-sized chunks this is a little // shift-and-add math. For variable-sized chunks this is a binary search, // with a sane API (JDK has an insane API). @Override public int elem2ChunkIdx(long i) { assert 0 <= i && i <= _len : " "+i+" < "+_len; int cidx = (int)(i/_chunkSize); int nc = nChunks(); if( i >= _len ) return nc; if( cidx >= nc ) cidx=nc-1; // Last chunk is larger assert 0 <= cidx && cidx < nc; return cidx; } // Convert a chunk-index into a starting row #. Constant sized chunks // (except for the last, which might be a little larger), and size-1 rows so // this is a little shift-n-add math. @Override long chunk2StartElem( int cidx ) { return (long)cidx*_chunkSize; } /** Convert a chunk-key to a file offset. Size 1-byte "rows", so this is a * direct conversion. * @return The file offset corresponding to this Chunk index */ public static long chunkOffset ( Key ckey ) { return (long)chunkIdx(ckey)*((FileVec)Vec.getVecKey(ckey).get())._chunkSize; } // Reverse: convert a chunk-key into a cidx static int chunkIdx(Key ckey) { assert ckey._kb[0]==Key.CHK; return UnsafeUtils.get4(ckey._kb, 1 + 1 + 4); } // Convert a chunk# into a chunk - does lazy-chunk creation. As chunks are // asked-for the first time, we make the Key and an empty backing DVec. // Touching the DVec will force the file load. @Override public Value chunkIdx( int cidx ) { final long nchk = nChunks(); assert 0 <= cidx && cidx < nchk; Key dkey = chunkKey(cidx); Value val1 = DKV.get(dkey);// Check for an existing one... will fetch data as needed if( val1 != null ) return val1; // Found an existing one? // Lazily create a DVec for this chunk int len = (int)(cidx < nchk-1 ? _chunkSize : (_len-chunk2StartElem(cidx))); // DVec is just the raw file data with a null-compression scheme Value val2 = new Value(dkey,len,null,TypeMap.C1NCHUNK,_be); val2.setDsk(); // It is already on disk. // If not-home, then block till the Key is everywhere. Most calls here are // from the parser loading a text file, and the parser splits the work such // that most puts here are on home - so this is a simple speed optimization: // do not make a Futures nor block on it on home. Futures fs = dkey.home() ? null : new Futures(); // Atomically insert: fails on a race, but then return the old version Value val3 = DKV.DputIfMatch(dkey,val2,null,fs); if( !dkey.home() && fs != null ) fs.blockForPending(); return val3 == null ? val2 : val3; } /** * Calculates safe and hopefully optimal chunk sizes. Four cases * exist. * <p> * very small data < 64K per core - uses default chunk size and * all data will be in one chunk * <p> * small data - data is partitioned into chunks that at least * 4 chunks per core to help keep all cores loaded * <p> * default - chunks are {@value #DFLT_CHUNK_SIZE} * <p> * large data - if the data would create more than 2M keys per * node, then chunk sizes larger than DFLT_CHUNK_SIZE are issued. * <p> * Too many keys can create enough overhead to blow out memory in * large data parsing. # keys = (parseSize / chunkSize) * numCols. * Key limit of 2M is a guessed "reasonable" number. * * @param totalSize - parse size in bytes (across all files to be parsed) * @param numCols - number of columns expected in dataset * @param cores - number of processing cores per node * @param cloudsize - number of compute nodes * @param verbose - print the parse heuristics * @return - optimal chunk size in bytes (always a power of 2). */ public static int calcOptimalChunkSize(long totalSize, int numCols, long maxLineLength, int cores, int cloudsize, boolean oldHeuristic, boolean verbose) { long localParseSize = (long) (double) totalSize / cloudsize; if (oldHeuristic) { long chunkSize = (localParseSize / (cores * 4)); // Super small data check - less than 64K/thread if (chunkSize <= (1 << 16)) { return DFLT_CHUNK_SIZE; } // Small data check chunkSize = 1L << MathUtils.log2(chunkSize); //closest power of 2 if (chunkSize < DFLT_CHUNK_SIZE && (localParseSize/chunkSize)*numCols < (1 << 21)) { // ignore if col cnt is high return (int)chunkSize; } // Big data check long tmp = (localParseSize * numCols / (1 << 21)); // ~ 2M keys per node if (tmp > (1 << 30)) return (1 << 30); // Max limit is 1G if (tmp > DFLT_CHUNK_SIZE) { chunkSize = 1 << MathUtils.log2((int) tmp); //closest power of 2 return (int)chunkSize; } else return DFLT_CHUNK_SIZE; } else { // New Heuristic int minNumberRows = 10; // need at least 10 rows (lines) per chunk (core) int perNodeChunkCountLimit = 1<<21; // don't create more than 2M Chunk POJOs per node int minParseChunkSize = 1<<12; // don't read less than this many bytes int maxParseChunkSize = (1<<28)-1; // don't read more than this many bytes per map() thread (needs to fit into a Value object) long chunkSize = Math.max((localParseSize / (4*cores))+1, minParseChunkSize); //lower hard limit if(chunkSize > 1024*1024) chunkSize = (chunkSize & 0xFFFFFE00) + 512; // align chunk size to 512B // Super small data check - file size is smaller than 64kB if (totalSize <= 1<<16) { chunkSize = Math.max(DFLT_CHUNK_SIZE, (int) (minNumberRows * maxLineLength)); } else { //round down to closest power of 2 // chunkSize = 1L << MathUtils.log2(chunkSize); // Small data check if (chunkSize < DFLT_CHUNK_SIZE && (localParseSize / chunkSize) * numCols < perNodeChunkCountLimit) { chunkSize = Math.max((int)chunkSize, (int) (minNumberRows * maxLineLength)); } else { // Adjust chunkSize such that we don't create too many chunks int chunkCount = cores * 4 * numCols; if (chunkCount > perNodeChunkCountLimit) { double ratio = 1 << Math.max(2, MathUtils.log2((int) (double) chunkCount / perNodeChunkCountLimit)); //this times too many chunks globally on the cluster chunkSize *= ratio; //need to bite off larger chunks } chunkSize = Math.min(maxParseChunkSize, chunkSize); // hard upper limit // if we can read at least minNumberRows and we don't create too large Chunk POJOs, we're done // else, fix it with a catch-all heuristic if (chunkSize <= minNumberRows * maxLineLength) { // might be more than default, if the max line length needs it, but no more than the size limit(s) // also, don't ever create too large chunks chunkSize = (int) Math.max( DFLT_CHUNK_SIZE, //default chunk size is a good lower limit for big data Math.min(maxParseChunkSize, minNumberRows * maxLineLength) //don't read more than 1GB, but enough to read the minimum number of rows ); } } } assert(chunkSize >= minParseChunkSize); assert(chunkSize <= maxParseChunkSize); if (verbose) Log.info("ParseSetup heuristic: " + "cloudSize: " + cloudsize + ", cores: " + cores + ", numCols: " + numCols + ", maxLineLength: " + maxLineLength + ", totalSize: " + totalSize + ", localParseSize: " + localParseSize + ", chunkSize: " + chunkSize + ", numChunks: " + Math.max(1,totalSize/chunkSize) + ", numChunks * cols: " + (Math.max(1,totalSize/chunkSize) * numCols) ); return (int)chunkSize; } } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/FileVecUtils.java
package water.fvec; import water.H2O; import water.Key; import water.Value; import water.persist.PersistManager; import java.io.IOException; import java.util.Map; public class FileVecUtils { public static byte[] getFirstBytes(FileVec vec) { return getFirstBytes(H2O.STORE, H2O.getPM(), vec); } static byte[] getFirstBytes(Map<Key, Value> store, PersistManager pm, FileVec vec) { if (store.get(vec.chunkKey(0)) != null) { // if it looks like we have the chunk cached attempt to use it instead of fetching it again return vec.getFirstChunkBytes(); } try { int max = (long) vec._chunkSize > vec._len ? (int) vec._len : vec._chunkSize; return pm.load(Value.HDFS, vec._key, 0L, max); } catch (IOException e) { throw new RuntimeException("HDFS read failed", e); } } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/Frame.java
package water.fvec; import org.joda.time.format.DateTimeFormat; import org.joda.time.format.DateTimeFormatter; import water.*; import water.api.FramesHandler; import water.api.schemas3.KeyV3; import water.exceptions.H2OFileAccessDeniedException; import water.exceptions.H2OIllegalArgumentException; import water.parser.BinaryFormatExporter; import water.parser.BufferedString; import water.rapids.Merge; import water.util.*; import java.io.IOException; import java.io.InputStream; import java.util.*; import java.util.regex.Pattern; /** A collection of named {@link Vec}s, essentially an R-like Distributed Data Frame. * * <p>Frames represent a large distributed 2-D table with named columns * ({@link Vec}s) and numbered rows. A reasonable <em>column</em> limit is * 100K columns, but there's no hard-coded limit. There's no real <em>row</em> * limit except memory; Frames (and Vecs) with many billions of rows are used * routinely. * * <p>A Frame is a collection of named Vecs; a Vec is a collection of numbered * {@link Chunk}s. A Frame is small, cheaply and easily manipulated, it is * commonly passed-by-Value. It exists on one node, and <em>may</em> be * stored in the {@link DKV}. Vecs, on the other hand, <em>must</em> be stored in the * {@link DKV}, as they represent the shared common management state for a collection * of distributed Chunks. * * <p>Multiple Frames can reference the same Vecs, although this sharing can * make Vec lifetime management complex. Commonly temporary Frames are used * to work with a subset of some other Frame (often during algorithm * execution, when some columns are dropped from the modeling process). The * temporary Frame can simply be ignored, allowing the normal GC process to * reclaim it. Such temp Frames usually have a {@code null} key. * * <p>All the Vecs in a Frame belong to the same {@link Vec.VectorGroup} which * then enforces {@link Chunk} row alignment across Vecs (or at least enforces * a low-cost access model). Parallel and distributed execution touching all * the data in a Frame relies on this alignment to get good performance. * * <p>Example: Make a Frame from a CSV file:<pre> * File file = ... * NFSFileVec nfs = NFSFileVec.make(file); // NFS-backed Vec, lazily read on demand * Frame fr = water.parser.ParseDataset.parse(Key.make("myKey"),nfs._key); * </pre> * * <p>Example: Find and remove the Vec called "unique_id" from the Frame, * since modeling with a unique_id can lead to overfitting: * <pre> * Vec uid = fr.remove("unique_id"); * </pre> * * <p>Example: Move the response column to the last position: * <pre> * fr.add("response",fr.remove("response")); * </pre> * */ public class Frame extends Lockable<Frame> { /** Vec names */ public String[] _names; private boolean _lastNameBig; // Last name is "Cxxx" and has largest number private Key<Vec>[] _keys; // Keys for the vectors private transient Vec[] _vecs; // The Vectors (transient to avoid network traffic) private transient Vec _col0; // First readable vec; fast access to the VectorGroup's Chunk layout /** * Given a temp Frame and a base Frame from which it was created, delete the * Vecs that aren't found in the base Frame and then delete the temp Frame. * * For most use cases with short-lived temp frames, use {@link Scope#protect(Frame...)} or a {@link Scope#safe(Frame...)} instead. */ public static void deleteTempFrameAndItsNonSharedVecs(Frame tempFrame, Frame baseFrame) { Key[] keys = tempFrame.keys(); for( int i=0; i<keys.length; i++ ) if( baseFrame.find(keys[i]) == -1 ) //only delete vecs that aren't shared Keyed.remove(keys[i]); if (tempFrame._key != null) DKV.remove(tempFrame._key); //delete the frame header } /** * Fetch all Frames from the KV store. */ public static Frame[] fetchAll() { // Get all the frames. final Key[] frameKeys = KeySnapshot.globalKeysOfClass(Frame.class); List<Frame> frames = new ArrayList<>(frameKeys.length); for( Key key : frameKeys ) { Frame frame = FramesHandler.getFromDKV("(none)", key); // Weed out frames with vecs that are no longer in DKV boolean skip = false; for( Vec vec : frame.vecs() ) { if (vec == null || DKV.get(vec._key) == null) { Log.warn("Leaked frame: Frame "+frame._key+" points to one or more deleted vecs."); skip = true; break; } } if (!skip) frames.add(frame); } return frames.toArray(new Frame[frames.size()]); } public boolean hasNAs(){ for(Vec v:bulkRollups()) if(v.naCnt() > 0) return true; return false; } public boolean hasInfs() { // return if frame contains positive infinity for (Vec v : bulkRollups()) if (v.pinfs() > 0 || v.ninfs() > 0) return true; return false; } private long _naCnt = -1; synchronized public long naCount() { if (_naCnt !=- 1) return _naCnt; _naCnt = 0; for(Vec v: vecs()) _naCnt += v.naCnt(); return _naCnt; } public double naFraction() { return naCount() / (numCols() * numRows()); } /** Creates an internal frame composed of the given Vecs and default names. The frame has no key. */ public Frame(Vec... vecs){ this((String[]) null, vecs); } /** Creates an internal frame composed of the given Vecs and names. The frame has no key. */ public Frame(String names[], Vec vecs[]) { this(null, names, vecs); } /** Creates an empty frame with given key. */ public Frame(Key<Frame> key) { this(key, null, new Vec[0]); } /** Creates a frame with given key, default names and vectors. */ public Frame(Key<Frame> key, Vec vecs[]) { this(key, null, vecs); } /** Creates a frame with given key, names and vectors. */ public Frame(Key<Frame> key, String names[], Vec vecs[] ) { super(key); // Require all Vecs already be installed in the K/V store for( Vec vec : vecs ) DKV.prefetch(vec._key); for( int i = 0; i < vecs.length; i++ ) { assert DKV.get(vecs[i]._key) != null : " null vec: " + vecs[i]._key + "; " + (names != null ? "name: " + names[i] : "index: " + i); } // Always require names if( names==null ) { // Make default names, all known to be unique setNames(new String[vecs.length]); _keys = makeVecKeys(vecs.length); _vecs = vecs; for( int i=0; i<vecs.length; i++ ) _names[i] = defaultColName(i); for( int i=0; i<vecs.length; i++ ) _keys [i] = vecs[i]._key; for( int i=0; i<vecs.length; i++ ) checkCompatibility(_names[i],vecs[i]); _lastNameBig = true; } else { // Make empty to dodge asserts, then "add()" them all which will check // for compatible Vecs & names. _names = new String[0]; _keys = makeVecKeys(0); _vecs = new Vec [0]; add(names,vecs); } assert _names.length == vecs.length; } void setNamesNoCheck(String[] columns){ _names = columns; } public final void setNames(String[] columns){ if (_vecs != null && columns.length != _vecs.length) { throw new IllegalArgumentException("Number of column names=" + columns.length + " must be the number of vecs=" + _vecs.length); } _names = columns; } /** Deep copy of Vecs and Keys and Names (but not data!) to a new random Key. * The resulting Frame does not share with the original, so the set of Vecs * can be freely hacked without disturbing the original Frame. */ public Frame( Frame fr ) { super( Key.<Frame>make() ); setNames(fr._names.clone()); _keys = fr._keys .clone(); _vecs = fr.vecs().clone(); _lastNameBig = fr._lastNameBig; } /** Default column name maker */ public static String defaultColName( int col ) { return "C"+(1+col); } /** * Helper method to initialize `_keys` array (which requires an unchecked cast). * @param size number of elements in the array that will be created. */ @SuppressWarnings("unchecked") private Key<Vec>[] makeVecKeys(int size) { return new Key[size]; } // Make unique names. Efficient for the special case of appending endless // versions of "C123" style names where the next name is +1 over the prior // name. All other names take the O(n^2) lookup. private int pint( String name ) { try { return Integer.valueOf(name.substring(1)); } catch(NumberFormatException ignored) { } return 0; } public String uniquify( String name ) { String n = name; int lastName = 0; if( name.length() > 0 && name.charAt(0)=='C' ) lastName = pint(name); if( _lastNameBig && _names.length > 0 ) { String last = _names[_names.length-1]; if( !last.equals("") && last.charAt(0)=='C' && lastName == pint(last)+1 ) return name; } int cnt=0, again, max=0; do { again = cnt; for( String s : _names ) { if( lastName > 0 && s.charAt(0)=='C' ) max = Math.max(max,pint(s)); if( n.equals(s) ) n = name+(cnt++); } } while( again != cnt ); if( lastName == max+1 ) _lastNameBig = true; return n; } /** Check that the vectors are all compatible. All Vecs have their content * sharded using same number of rows per chunk, and all names are unique. * Throw an IAE if something does not match. */ private void checkCompatibility(String name, Vec vec ) { if( vec instanceof AppendableVec ) return; // New Vectors are endlessly compatible Vec v0 = anyVec(); if( v0 == null ) return; // No fixed-size Vecs in the Frame // Vector group has to be the same, or else the layout has to be the same, // or else the total length has to be small. if( !v0.isCompatibleWith(vec) ) { if(!Vec.VectorGroup.sameGroup(v0,vec)) Log.err("Unexpected incompatible vector group, " + v0.group() + " != " + vec.group()); if(!Arrays.equals(v0.espc(), vec.espc())) Log.err("Unexpected incompatible espc, " + Arrays.toString(v0.espc()) + " != " + Arrays.toString(vec.espc())); throw new IllegalArgumentException("Vec " + name + " is not compatible with the rest of the frame"); } } /** Frames are compatible if they have the same layout (number of rows and chunking) and the same vector group (chunk placement).. */ public boolean isCompatible( Frame fr ) { if( numRows() != fr.numRows() ) return false; for( int i=0; i<vecs().length; i++ ) if( !vecs()[i].isCompatibleWith(fr.vecs()[i]) ) return false; return true; } /** Number of columns * @return Number of columns */ public int numCols() { return _keys == null? 0 : _keys.length; } /** Number of rows * @return Number of rows */ public long numRows() { Vec v = anyVec(); return v==null ? 0 : v.length(); } /** Returns the first readable vector. * @return the first readable Vec */ public final Vec anyVec() { Vec c0 = _col0; // single read if( c0 != null ) return c0; for( Vec v : vecs() ) if( v.readable() ) return (_col0 = v); return null; } /** The array of column names. * @return the array of column names */ public String[] names() { return _names; } /** A single column name. * @return the column name */ public String name(int i) { return _names[i]; } /** The array of keys. * @return the array of keys for each vec in the frame. */ public Key<Vec>[] keys() { return _keys; } public Iterable<Key<Vec>> keysList() { return Arrays.asList(_keys); } /** The internal array of Vecs. For efficiency Frames contain an array of * Vec Keys - and the Vecs themselves are lazily loaded from the {@link DKV}. * @return the internal array of Vecs */ public final Vec[] vecs() { Vec[] tvecs = _vecs; // read the content return tvecs == null ? (_vecs=vecs_impl()) : tvecs; } public final Vec[] vecs(int [] idxs) { Vec [] all = vecs(); Vec [] res = new Vec[idxs.length]; for(int i = 0; i < idxs.length; ++i) res[i] = all[idxs[i]]; return res; } public Vec[] vecs(String[] names) { Vec [] res = new Vec[names.length]; for(int i = 0; i < names.length; ++i) res[i] = vec(names[i]); return res; } // Compute vectors for caching private Vec[] vecs_impl() { // Load all Vec headers; load them all in parallel by starting prefetches for( Key<Vec> key : _keys ) DKV.prefetch(key); Vec [] vecs = new Vec[_keys.length]; for( int i=0; i<_keys.length; i++ ) vecs[i] = _keys[i].get(); return vecs; } /** Convenience to accessor for last Vec * @return last Vec */ public Vec lastVec() { vecs(); return _vecs [_vecs.length -1]; } /** Convenience to accessor for last Vec name * @return last Vec name */ public String lastVecName() { return _names[_names.length-1]; } /** Force a cache-flush and reload, assuming vec mappings were altered * remotely, or that the _vecs array was shared and now needs to be a * defensive copy. * @return the new instance of the Frame's Vec[] */ public final Vec[] reloadVecs() { _vecs=null; return vecs(); } /** Returns the Vec by given index, implemented by code: {@code vecs()[idx]}. * @param idx idx of column * @return this frame idx-th vector, never returns <code>null</code> */ public final Vec vec(int idx) { return vecs()[idx]; } /** Return a Vec by name, or null if missing * @return a Vec by name, or null if missing */ public Vec vec(String name) { int idx = find(name); return idx==-1 ? null : vecs()[idx]; } /** Finds the column index with a matching name, or -1 if missing * @return the column index with a matching name, or -1 if missing */ public int find( String name ) { if( name == null ) return -1; assert _names != null; // TODO: add a hashtable: O(n) is just stupid. for( int i=0; i<_names.length; i++ ) if( name.equals(_names[i]) ) return i; return -1; } /** Finds the matching column index, or -1 if missing * @return the matching column index, or -1 if missing * @deprecated as many columns in a Frame could be backed by the same Vec, we can't return single column index that corresponds to a given {@code vec}. * Please use {@link #find(String)} instead. */ @Deprecated() public int find( Vec vec ) { Vec[] vecs = vecs(); //warning: side-effect if (vec == null) return -1; for( int i=0; i<vecs.length; i++ ) if( vec.equals(vecs[i]) ) return i; return -1; } /** Finds the matching column index, or -1 if missing * @return the matching column index, or -1 if missing * @deprecated as many columns in a Frame could be backed by the same Vec (and its key), we can't return single column index that corresponds to a given {@code key}. * Please use {@link #find(String)} instead. */ @Deprecated public int find( Key key ) { for( int i=0; i<_keys.length; i++ ) if( key.equals(_keys[i]) ) return i; return -1; } /** Bulk {@link #find(String)} api * @return An array of column indices matching the {@code names} array */ public int[] find(String[] names) { if( names == null ) return null; int[] res = new int[names.length]; for(int i = 0; i < names.length; ++i) res[i] = find(names[i]); return res; } public void insertVec(int i, String name, Vec vec) { String [] names = new String[_names.length+1]; Vec [] vecs = new Vec[_vecs.length+1]; Key<Vec>[] keys = makeVecKeys(_keys.length + 1); System.arraycopy(_names,0,names,0,i); System.arraycopy(_vecs,0,vecs,0,i); System.arraycopy(_keys,0,keys,0,i); names[i] = name; vecs[i] = vec; keys[i] = vec._key; System.arraycopy(_names,i,names,i+1,_names.length-i); System.arraycopy(_vecs,i,vecs,i+1,_vecs.length-i); System.arraycopy(_keys,i,keys,i+1,_keys.length-i); _vecs = vecs; setNames(names); _keys = keys; } /** Pair of (column name, Frame key). */ public static class VecSpecifier extends Iced implements Vec.Holder { public Key<Frame> _frame; public String _column_name; public VecSpecifier() { } public VecSpecifier(Key<Frame> frame, String column_name) { _frame = frame; _column_name = column_name; } public Vec vec() { Value v = DKV.get(_frame); if (null == v) return null; Frame f = v.get(); if (null == f) return null; return f.vec(_column_name); } /** * @param vecSpecifiers An Array of vectors specifiers to extract column names from. May not be null. * @return An array of String with names of the columns represented by each given VecSpecifier. Possibly empty. * @throws NullPointerException When vecSpecifiers argument is null. */ public static String[] vecNames(final VecSpecifier[] vecSpecifiers) throws NullPointerException { Objects.requireNonNull(vecSpecifiers); final String[] vecNames = new String[vecSpecifiers.length]; for (int i = 0; i < vecSpecifiers.length; i++) { vecNames[i] = vecSpecifiers[i]._column_name; } return vecNames; } } /** Type for every Vec */ public byte[] types() { Vec[] vecs = vecs(); byte bs[] = new byte[vecs.length]; for( int i=0; i<vecs.length; i++ ) bs[i] = vecs[i]._type; return bs; } /** String name for each Vec type */ public String[] typesStr() { // typesStr not strTypes since shows up in intelliJ next to types Vec[] vecs = vecs(); String s[] = new String[vecs.length]; for(int i=0;i<vecs.length;++i) s[i] = vecs[i].get_type_str(); return s; } /** All the domains for categorical columns; null for non-categorical columns. * @return the domains for categorical columns */ public String[][] domains() { Vec[] vecs = vecs(); String ds[][] = new String[vecs.length][]; for( int i=0; i<vecs.length; i++ ) { ds[i] = vecs[i].domain(); } return ds; } /** Number of categorical levels for categorical columns; -1 for non-categorical columns. * @return the number of levels for categorical columns */ public int[] cardinality() { Vec[] vecs = vecs(); int[] card = new int[vecs.length]; for( int i=0; i<vecs.length; i++ ) card[i] = vecs[i].cardinality(); return card; } public Vec[] bulkRollups() { Futures fs = new Futures(); Vec[] vecs = vecs(); for(Vec v : vecs) v.startRollupStats(fs); fs.blockForPending(); return vecs; } /** Majority class for categorical columns; -1 for non-categorical columns. * @return the majority class for categorical columns */ public int[] modes() { Vec[] vecs = bulkRollups(); int[] modes = new int[vecs.length]; for( int i = 0; i < vecs.length; i++ ) { modes[i] = vecs[i].isCategorical() ? vecs[i].mode() : -1; } return modes; } /** All the column means. * @return the mean of each column */ public double[] means() { Vec[] vecs = bulkRollups(); double[] means = new double[vecs.length]; for( int i = 0; i < vecs.length; i++ ) means[i] = vecs[i].mean(); return means; } /** One over the standard deviation of each column. * @return Reciprocal the standard deviation of each column */ public double[] mults() { Vec[] vecs = bulkRollups(); double[] mults = new double[vecs.length]; for( int i = 0; i < vecs.length; i++ ) { double sigma = vecs[i].sigma(); mults[i] = standardize(sigma) ? 1.0 / sigma : 1.0; } return mults; } private static boolean standardize(double sigma) { // TODO unify handling of constant columns return sigma > 1e-6; } /** The {@code Vec.byteSize} of all Vecs * @return the {@code Vec.byteSize} of all Vecs */ public long byteSize() { try { Vec[] vecs = bulkRollups(); long sum = 0; for (Vec vec : vecs) sum += vec.byteSize(); return sum; } catch(RuntimeException ex) { Log.debug("Failure to obtain byteSize() - missing chunks?"); return -1; } } /** 64-bit checksum of the checksums of the vecs. SHA-265 checksums of the * chunks are XORed together. Since parse always parses the same pieces of * files into the same offsets in some chunk this checksum will be * consistent across reparses. * @return 64-bit Frame checksum */ @Override protected long checksum_impl(boolean noCache) { Vec[] vecs = vecs(); long _checksum = 0; for( int i = 0; i < _names.length; ++i ) { long vec_checksum = vecs[i].checksum(noCache); _checksum ^= vec_checksum; long tmp = (2147483647L * i); _checksum ^= tmp; } _checksum *= (0xBABE + Arrays.hashCode(_names)); // TODO: include column types? Vec.checksum() should include type? return _checksum; } // Add a bunch of vecs public void add( String[] names, Vec[] vecs) { bulkAdd(names, vecs); } public void add( String[] names, Vec[] vecs, int cols ) { if (null == vecs || null == names) return; if (cols == names.length && cols == vecs.length) { bulkAdd(names, vecs); } else { for (int i = 0; i < cols; i++) add(names[i], vecs[i]); } } /** Append multiple named Vecs to the Frame. Names are forced unique, by appending a * unique number if needed. */ private void bulkAdd(String[] names, Vec[] vecs) { String[] tmpnames = names.clone(); int N = names.length; assert(names.length == vecs.length):"names = " + Arrays.toString(names) + ", vecs len = " + vecs.length; for (int i=0; i<N; ++i) { vecs[i] = vecs[i] != null ? makeCompatible(new Frame(vecs[i]))[0] : null; checkCompatibility(tmpnames[i]=uniquify(tmpnames[i]),vecs[i]); // Throw IAE is mismatch } int ncols = _keys.length; // make temp arrays and don't assign them back until they are fully filled - otherwise vecs() can cache null's and NPE. String[] tmpnam = Arrays.copyOf(_names, ncols+N); Key<Vec>[] tmpkeys = Arrays.copyOf(_keys, ncols+N); Vec[] tmpvecs = Arrays.copyOf(_vecs, ncols+N); for (int i=0; i<N; ++i) { tmpnam[ncols+i] = tmpnames[i]; tmpkeys[ncols+i] = vecs[i]._key; tmpvecs[ncols+i] = vecs[i]; } _keys = tmpkeys; _vecs = tmpvecs; setNames(tmpnam); } /** Append a named Vec to the Frame. Names are forced unique, by appending a * unique number if needed. * @return the added Vec, for flow-coding */ public Vec add( String name, Vec vec ) { vec = makeCompatible(new Frame(vec))[0]; checkCompatibility(name=uniquify(name),vec); // Throw IAE is mismatch int ncols = _keys.length; String[] names = Arrays.copyOf(_names,ncols+1); names[ncols] = name; Key<Vec>[] keys = Arrays.copyOf(_keys ,ncols+1); keys [ncols] = vec._key; Vec[] vecs = Arrays.copyOf(_vecs ,ncols+1); vecs [ncols] = vec; _keys = keys; _vecs = vecs; setNames(names); return vec; } /** Append a Frame onto this Frame. Names are forced unique, by appending * unique numbers if needed. * @return the expanded Frame, for flow-coding */ public Frame add( Frame fr ) { add(fr._names,fr.vecs().clone(),fr.numCols()); return this; } /** Insert a named column as the first column */ public Frame prepend( String name, Vec vec ) { if( find(name) != -1 ) throw new IllegalArgumentException("Duplicate name '"+name+"' in Frame"); if( _vecs.length != 0 ) { if( !anyVec().group().equals(vec.group()) && !Arrays.equals(anyVec().espc(),vec.espc()) ) throw new IllegalArgumentException("Vector groups differs - adding vec '"+name+"' into the frame " + Arrays.toString(_names)); if( numRows() != vec.length() ) throw new IllegalArgumentException("Vector lengths differ - adding vec '"+name+"' into the frame " + Arrays.toString(_names)); } final int len = _names != null ? _names.length : 0; String[] _names2 = new String[len + 1]; Vec[] _vecs2 = new Vec[len + 1]; Key<Vec>[] _keys2 = makeVecKeys(len + 1); _names2[0] = name; _vecs2 [0] = vec; _keys2 [0] = vec._key; if (_names != null) { System.arraycopy(_names, 0, _names2, 1, len); System.arraycopy(_vecs, 0, _vecs2, 1, len); System.arraycopy(_keys, 0, _keys2, 1, len); } _vecs = _vecs2; _keys = _keys2; setNames(_names2); return this; } /** Swap two Vecs in-place; useful for sorting columns by some criteria */ public void swap( int lo, int hi ) { assert 0 <= lo && lo < _keys.length; assert 0 <= hi && hi < _keys.length; if( lo==hi ) return; Vec vecs[] = vecs(); Vec v = vecs [lo]; vecs [lo] = vecs [hi]; vecs [hi] = v; Key<Vec> k = _keys[lo]; _keys[lo] = _keys[hi]; _keys[hi] = k; String n=_names[lo]; _names[lo] = _names[hi]; _names[hi] = n; } /** * Re-order the columns according to the new order specified in newOrder. * * @param newOrder */ public void reOrder(int[] newOrder) { assert newOrder.length==_keys.length; // make sure column length match int numCols = _keys.length; Vec tmpvecs[] = vecs().clone(); Key<Vec> tmpkeys[] = _keys.clone(); String tmpnames[] = _names.clone(); for (int colIndex = 0; colIndex < numCols; colIndex++) { tmpvecs[colIndex] = _vecs[newOrder[colIndex]]; tmpkeys[colIndex] = _keys[newOrder[colIndex]]; tmpnames[colIndex] = _names[newOrder[colIndex]]; } // copy it back for (int colIndex = 0; colIndex < numCols; colIndex++) { _vecs[colIndex] = tmpvecs[colIndex]; _keys[colIndex] = tmpkeys[colIndex]; _names[colIndex] = tmpnames[colIndex]; } } /** move the provided columns to be first, in-place. For Merge currently since method='hash' was coded like that */ public void moveFirst( int cols[] ) { boolean colsMoved[] = new boolean[_keys.length]; Vec tmpvecs[] = vecs().clone(); Key<Vec> tmpkeys[] = _keys.clone(); String tmpnames[] = _names.clone(); // Move the desired ones first for (int i=0; i<cols.length; i++) { int w = cols[i]; if (colsMoved[w]) throw new IllegalArgumentException("Duplicates in column numbers passed in"); if (w<0 || w>=_keys.length) throw new IllegalArgumentException("column number out of 0-based range"); colsMoved[w] = true; tmpvecs[i] = _vecs[w]; tmpkeys[i] = _keys[w]; tmpnames[i] = _names[w]; } // Put the other ones afterwards int w = cols.length; for (int i=0; i<_keys.length; i++) { if (!colsMoved[i]) { tmpvecs[w] = _vecs[i]; tmpkeys[w] = _keys[i]; tmpnames[w] = _names[i]; w++; } } // Copy back over the original in-place for (int i=0; i<_keys.length; i++) { _vecs[i] = tmpvecs[i]; _keys[i] = tmpkeys[i]; _names[i] = tmpnames[i]; } } /** Returns a subframe of this frame containing only vectors with desired names. * * @param names list of vector names * @return a new frame which collects vectors from this frame with desired names. * @throws IllegalArgumentException if there is no vector with desired name in this frame. */ public Frame subframe(String[] names) { Vec[] vecs = new Vec[names.length]; vecs(); // Preload the vecs HashMap<String, Integer> map = new HashMap<>((int) ((names.length/0.75f)+1)); // avoid rehashing by set up initial capacity for (int i = 0; i < _names.length; i++) map.put(_names[i], i); int missingCnt = 0; for (int i = 0; i < names.length; i++) if (map.containsKey(names[i])) { vecs[i] = _vecs[map.get(names[i])]; } else missingCnt++; if (missingCnt > 0) { StringBuilder sb = new StringBuilder(); final int maxReported = missingCnt <= 5 ? missingCnt : 5; int reported = 0; for (int i = 0; i < names.length && reported < maxReported; i++) { if (vecs[i] == null) { sb.append('\'').append(names[i]).append('\''); reported++; if (reported < maxReported) { sb.append(", "); } } } if (reported < missingCnt) { sb.append(" (and other ").append(missingCnt - reported).append(")"); } throw new IllegalArgumentException("Frame `" + _key + "` doesn't contain columns: " + sb.toString() + "."); } return new Frame(Key.<Frame>make("subframe" + Key.make().toString()), names, vecs); } /** Allow rollups for all written-into vecs; used by {@link MRTask} once * writing is complete. * @return the original Futures, for flow-coding */ public Futures postWrite(Futures fs) { for( Vec v : vecs() ) v.postWrite(fs); return fs; } /** Actually remove/delete all Vecs from memory, not just from the Frame. * @return the original Futures, for flow-coding */ @Override protected Futures remove_impl(Futures fs, boolean cascade) { final Key<Vec>[] keys = _keys; if( keys.length==0 ) return fs; // Get the nChunks without calling anyVec - which loads all Vecs eagerly, // only to delete them. Supports Frames with some Vecs already deleted, as // a Scope cleanup action might delete Vecs out of order. Vec v = _col0; if (v == null) { Vec[] vecs = _vecs; // Read once, in case racily being cleared if (vecs != null) for (Vec vec : vecs) if ((v = vec) != null) // Stop on finding the 1st Vec break; } if (v == null) // Ok, now do DKV gets for (Key<Vec> _key1 : _keys) if ((v = _key1.get()) != null) break; // Stop on finding the 1st Vec if (v == null) return fs; _vecs = new Vec[0]; setNames(new String[0]); _keys = makeVecKeys(0); if (cascade) { // removing the vecs from mem only if cascading (default behaviour) // Bulk dumb local remove - no JMM, no ordering, no safety. Vec.bulk_remove(keys, v.nChunks()); } return fs; } @Override public Frame delete_and_lock(Key<Job> job_key) { return super.delete_and_lock(job_key, true); // for Frames, remove dependencies (Vecs) by default when forcing internal delete } /** * Removes this {@link Frame} object and all directly linked {@link Keyed} objects and POJOs, while retaining * the keys defined by the retainedKeys parameter. Aimed to be used for removal of {@link Frame} objects pointing * to shared resources (Vectors, Chunks etc.) internally. * <p> * WARNING: UNSTABLE API, might be removed/replaced at any time. * * @param futures An instance of {@link Futures} for synchronization * @param retainedKeys A {@link Set} of keys to retain. The set may be immutable, as it shall not be modified. * @return An instance of {@link Futures} for synchronization */ public final Futures retain(final Futures futures, final Set<Key> retainedKeys) { if (_key != null) DKV.remove(_key); final Key[] delCandidateKeys = _keys; if (delCandidateKeys.length == 0) return futures; // Get the nChunks without calling anyVec - which loads all Vecs eagerly, // only to delete them. Supports Frames with some Vecs already deleted, as // a Scope cleanup action might delete Vecs out of order. Vec v = _col0; if (v == null) { Vec[] vecs = _vecs; // Read once, in case racily being cleared if (vecs != null) for (Vec vec : vecs) if ((v = vec) != null) // Stop on finding the 1st Vec break; } if (v == null) // Ok, now do DKV gets for (Key<Vec> _key1 : _keys) if ((v = _key1.get()) != null) break; // Stop on finding the 1st Vec if (v == null) return futures; _vecs = new Vec[0]; setNames(new String[0]); _keys = makeVecKeys(0); final List<Key> deletedKeys= new ArrayList<>(); for (int i = 0; i < delCandidateKeys.length; i++) { if(!retainedKeys.contains(delCandidateKeys[i])){ deletedKeys.add(delCandidateKeys[i]); } } // Bulk dumb local remove - no JMM, no ordering, no safety. Vec.bulk_remove(deletedKeys.toArray(new Key[]{}), v.nChunks()); return futures; } /** Write out K/V pairs, in this case Vecs. */ @Override protected AutoBuffer writeAll_impl(AutoBuffer ab) { ab.putA8(anyVec().espc()); for (Key k : _keys) ab.putKey(k); return super.writeAll_impl(ab); } @Override protected Keyed readAll_impl(AutoBuffer ab, Futures fs) { long[] espc = ab.getA8(); _keys = new Vec.VectorGroup().addVecs(_keys.length); // I am modifying self => I need to make an update // This is more of a workaround, readAll_impl methods are not expected to modify self DKV.put(this, fs); int rowLayout = Vec.ESPC.rowLayout(_keys[0], espc); for (Key<Vec> key : _keys) { Vec v = ab.get(); v._key = key; v._rowLayout = rowLayout; v.readAll_impl(ab, fs); DKV.put(v, fs); } return super.readAll_impl(ab,fs); } /** Replace one column with another. Caller must perform global update (DKV.put) on * this updated frame. * @return The old column, for flow-coding */ public Vec replace(int col, Vec nv) { Vec rv = vecs()[col]; nv = ((new Frame(rv)).makeCompatible(new Frame(nv)))[0]; DKV.put(nv); assert DKV.get(nv._key)!=null; // Already in DKV assert rv.isCompatibleWith(nv); _vecs[col] = nv; _keys[col] = nv._key; return rv; } /** Create a subframe from given interval of columns. * @param startIdx index of first column (inclusive) * @param endIdx index of the last column (exclusive) * @return a new Frame containing specified interval of columns */ public Frame subframe(int startIdx, int endIdx) { return new Frame(Arrays.copyOfRange(_names,startIdx,endIdx),Arrays.copyOfRange(vecs(),startIdx,endIdx)); } /** Split this Frame; return a subframe created from the given column interval, and * remove those columns from this Frame. * @param startIdx index of first column (inclusive) * @param endIdx index of the last column (exclusive) * @return a new Frame containing specified interval of columns */ public Frame extractFrame(int startIdx, int endIdx) { Frame f = subframe(startIdx, endIdx); remove(startIdx, endIdx); return f; } /** Removes the column with a matching name. * @return The removed column */ public Vec remove( String name ) { return remove(find(name)); } public Frame remove( String[] names ) { for( String name : names ) remove(find(name)); return this; } /** Removes a list of columns by index; the index list must be sorted * @return an array of the removed columns */ public Vec[] remove( int[] idxs ) { for( int i : idxs ) if(i < 0 || i >= vecs().length) throw new ArrayIndexOutOfBoundsException(); Arrays.sort(idxs); Vec[] res = new Vec[idxs.length]; Vec[] rem = new Vec[_vecs.length-idxs.length]; String[] names = new String[rem.length]; Key<Vec>[] keys = makeVecKeys(rem.length); int j = 0; int k = 0; int l = 0; for(int i = 0; i < _vecs.length; ++i) { if(j < idxs.length && i == idxs[j]) { ++j; res[k++] = _vecs[i]; } else { rem [l] = _vecs [i]; names[l] = _names[i]; keys [l] = _keys [i]; ++l; } } _vecs = rem; setNames(names); _keys = keys; assert l == rem.length && k == idxs.length; return res; } /** Removes a numbered column. * @return the removed column */ public final Vec remove( int idx ) { int len = _names.length; if( idx < 0 || idx >= len ) return null; Vec v = vecs()[idx]; if( v == _col0 ) _col0 = null; _vecs = ArrayUtils.remove(_vecs, idx); setNames(ArrayUtils.remove(_names, idx)); _keys = ArrayUtils.remove(_keys, idx); return v; } /** * Remove all the vecs from frame. */ public Vec[] removeAll() { return remove(0, _names.length); } /** Remove given interval of columns from frame. Motivated by R intervals. * @param startIdx - start index of column (inclusive) * @param endIdx - end index of column (exclusive) * @return array of removed columns */ Vec[] remove(int startIdx, int endIdx) { int len = _names.length; int nlen = len - (endIdx-startIdx); String[] names = new String[nlen]; Key<Vec>[] keys = makeVecKeys(nlen); Vec[] vecs = new Vec[nlen]; vecs(); if (startIdx > 0) { System.arraycopy(_names, 0, names, 0, startIdx); System.arraycopy(_vecs, 0, vecs, 0, startIdx); System.arraycopy(_keys, 0, keys, 0, startIdx); } nlen -= startIdx; if (endIdx < _names.length+1) { System.arraycopy(_names, endIdx, names, startIdx, nlen); System.arraycopy(_vecs, endIdx, vecs, startIdx, nlen); System.arraycopy(_keys, endIdx, keys, startIdx, nlen); } Vec[] vecX = Arrays.copyOfRange(_vecs,startIdx,endIdx); _vecs = vecs; _keys = keys; setNames(names); _col0 = null; return vecX; } /** Restructure a Frame completely */ public void restructure( String[] names, Vec[] vecs) { restructure(names, vecs, vecs.length); } /** Restructure a Frame completely, but only for a specified number of columns (counting up) */ public void restructure( String[] names, Vec[] vecs, int cols) { // Make empty to dodge asserts, then "add()" them all which will check for // compatible Vecs & names. _keys = makeVecKeys(0); _vecs = new Vec [0]; setNames(new String[0]); add(names,vecs,cols); } // -------------------------------------------- // Utilities to help external Frame constructors, e.g. Spark. // Make an initial Frame & lock it for writing. Build Vec Keys. void preparePartialFrame( String[] names ) { // Nuke any prior frame (including freeing storage) & lock this one if( _keys != null ) delete_and_lock(); else write_lock(); _keys = new Vec.VectorGroup().addVecs(names.length); setNamesNoCheck(names); // No Vectors tho!!! These will be added *after* the import } // Only serialize strings, not H2O internal structures // Make NewChunks to for holding data from e.g. Spark. Once per set of // Chunks in a Frame, before filling them. This can be called in parallel // for different Chunk#'s (cidx); each Chunk can be filled in parallel. static NewChunk[] createNewChunks(String name, byte[] type, int cidx) { boolean[] sparse = new boolean[type.length]; Arrays.fill(sparse, false); return createNewChunks(name, type, cidx, sparse); } static NewChunk[] createNewChunks(String name, byte[] type, int cidx, boolean[] sparse) { Frame fr = (Frame) Key.make(name).get(); NewChunk[] nchks = new NewChunk[fr.numCols()]; for (int i = 0; i < nchks.length; i++) { nchks[i] = new NewChunk(new AppendableVec(fr._keys[i], type[i]), cidx, sparse[i]); } return nchks; } // Compress & DKV.put NewChunks. Once per set of Chunks in a Frame, after // filling them. Can be called in parallel for different sets of Chunks. static void closeNewChunks(NewChunk[] nchks) { Futures fs = new Futures(); for (NewChunk nchk : nchks) { nchk.close(fs); } fs.blockForPending(); } // Build real Vecs from loose Chunks, and finalize this Frame. Called once // after any number of [create,close]NewChunks. This method also unlocks the frame. void finalizePartialFrame(long[] espc, String[][] domains, byte[] types) { finalizePartialFrame(espc, domains, types, true); } // Build real Vecs from loose Chunks, and finalize this Frame. Called once // after any number of [create,close]NewChunks. void finalizePartialFrame(long[] espc, String[][] domains, byte[] types, boolean unlock) { // Compute elems-per-chunk. // Roll-up elem counts, so espc[i] is the starting element# of chunk i. int nchunk = espc.length; while( nchunk > 1 && espc[nchunk-1] == 0 ) nchunk--; long espc2[] = new long[nchunk+1]; // Shorter array long x=0; // Total row count so far for( int i=0; i<nchunk; i++ ) { espc2[i] = x; // Start elem# for chunk i x += espc[i]; // Raise total elem count } espc2[nchunk]=x; // Total element count in last // For all Key/Vecs - insert Vec header Futures fs = new Futures(); _vecs = new Vec[_keys.length]; for( int i=0; i<_keys.length; i++ ) { // Nuke the extra chunks for (int j = nchunk; j < espc.length; j++) DKV.remove(Vec.chunkKey(_keys[i], j), fs); // Insert Vec header Vec vec = _vecs[i] = new Vec( _keys[i], Vec.ESPC.rowLayout(_keys[i],espc2), domains!=null ? domains[i] : null, types[i]); // Here we have to save vectors since // saving during unlock will invoke Frame vector // refresh DKV.put(_keys[i],vec,fs); } fs.blockForPending(); if (unlock) { unlock(); } } // -------------------------------------------------------------------------- static final int MAX_EQ2_COLS = 100000; // Limit of columns user is allowed to request /** In support of R, a generic Deep Copy and Slice. * * <p>Semantics are a little odd, to match R's. Each dimension spec can be:<ul> * <li><em>null</em> - all of them * <li><em>a sorted list of negative numbers (no dups)</em> - all BUT these * <li><em>an unordered list of positive</em> - just these, allowing dups * </ul> * * <p>The numbering is 1-based; zero's are not allowed in the lists, nor are out-of-range values. * @return the sliced Frame */ public Frame deepSlice( Object orows, Object ocols ) { // ocols is either a long[] or a Frame-of-1-Vec long[] cols; if( ocols == null ) cols = null; else if (ocols instanceof long[]) cols = (long[])ocols; else if (ocols instanceof Frame) { Frame fr = (Frame) ocols; if (fr.numCols() != 1) throw new IllegalArgumentException("Columns Frame must have only one column (actually has " + fr.numCols() + " columns)"); long n = fr.anyVec().length(); if (n > MAX_EQ2_COLS) throw new IllegalArgumentException("Too many requested columns (requested " + n +", max " + MAX_EQ2_COLS + ")"); cols = new long[(int)n]; Vec.Reader v = fr.anyVec().new Reader(); for (long i = 0; i < v.length(); i++) cols[(int)i] = v.at8(i); } else throw new IllegalArgumentException("Columns is specified by an unsupported data type (" + ocols.getClass().getName() + ")"); // Since cols is probably short convert to a positive list. int c2[]; if( cols==null ) { c2 = new int[numCols()]; for( int i=0; i<c2.length; i++ ) c2[i]=i; } else if( cols.length==0 ) { c2 = new int[0]; } else if( cols[0] >= 0 ) { c2 = new int[cols.length]; for( int i=0; i<cols.length; i++ ) c2[i] = (int)cols[i]; // Conversion of 1-based cols to 0-based is handled by a 1-based front-end! } else { c2 = new int[numCols()-cols.length]; int j=0; for( int i=0; i<numCols(); i++ ) { if( j >= cols.length || i < (-(1+cols[j])) ) c2[i-j] = i; else j++; } } for (int aC2 : c2) if (aC2 >= numCols()) throw new IllegalArgumentException("Trying to select column " + (aC2 + 1) + " but only " + numCols() + " present."); if( c2.length==0 ) throw new IllegalArgumentException("No columns selected (did you try to select column 0 instead of column 1?)"); // Do Da Slice // orows is either a long[] or a Vec if (numRows() == 0) { return new MRTask() { @Override public void map(Chunk[] chks, NewChunk[] nchks) { for (NewChunk nc : nchks) nc.addNA(); } }.doAll(types(c2), this).outputFrame(names(c2), domains(c2)); } if (orows == null) return new DeepSlice(null,c2,vecs()).doAll(types(c2),this).outputFrame(names(c2),domains(c2)); else if (orows instanceof long[]) { final long CHK_ROWS=1000000; final long[] rows = (long[])orows; if (this.numRows() == 0) { return this; } if( rows.length==0 || rows[0] < 0 ) { if (rows.length != 0 && rows[0] < 0) { Vec v0 = this.anyVec().makeZero(); Vec v = new MRTask() { @Override public void map(Chunk cs) { for (long er : rows) { if (er >= 0) continue; er = Math.abs(er); if (er < cs._start || er > (cs._len + cs._start - 1)) continue; cs.set((int) (er - cs._start), 1); } } }.doAll(v0).getResult()._fr.anyVec(); Keyed.remove(v0._key); Frame slicedFrame = new DeepSlice(rows, c2, vecs()).doAll(types(c2), this.add("select_vec", v)).outputFrame(names(c2), domains(c2)); Keyed.remove(v._key); Keyed.remove(this.remove(this.numCols() - 1)._key); return slicedFrame; } else { return new DeepSlice(rows.length == 0 ? null : rows, c2, vecs()).doAll(types(c2), this).outputFrame(names(c2), domains(c2)); } } // Vec'ize the index array Futures fs = new Futures(); AppendableVec av = new AppendableVec(Vec.newKey(),Vec.T_NUM); int r = 0; int c = 0; while (r < rows.length) { NewChunk nc = new NewChunk(av, c); long end = Math.min(r+CHK_ROWS, rows.length); for (; r < end; r++) { nc.addNum(rows[r]); } nc.close(c++, fs); } Vec c0 = av.layout_and_close(fs); // c0 is the row index vec fs.blockForPending(); Frame ff = new Frame(new String[]{"rownames"}, new Vec[]{c0}); Frame fr2 = new Slice(c2, this).doAll(types(c2),ff).outputFrame(names(c2), domains(c2)); Keyed.remove(c0._key); Keyed.remove(av._key); ff.delete(); return fr2; } Frame frows = (Frame)orows; // It's a compatible Vec; use it as boolean selector. // Build column names for the result. Vec [] vecs = new Vec[c2.length]; String [] names = new String[c2.length]; for(int i = 0; i < c2.length; ++i){ vecs[i] = _vecs[c2[i]]; names[i] = _names[c2[i]]; } Frame ff = new Frame(names, vecs); ff.add("predicate", frows.anyVec()); return new DeepSelect().doAll(types(c2),ff).outputFrame(names(c2),domains(c2)); } // Slice and return in the form of new chunks. private static class Slice extends MRTask<Slice> { final Frame _base; // the base frame to slice from final int[] _cols; Slice(int[] cols, Frame base) { _cols = cols; _base = base; } @Override public void map(Chunk[] ix, NewChunk[] ncs) { final Vec[] vecs = new Vec[_cols.length]; final Vec anyv = _base.anyVec(); final long nrow = anyv.length(); long r = ix[0].at8(0); int last_ci = anyv.elem2ChunkIdx(r<nrow?r:0); // memoize the last chunk index long last_c0 = anyv.espc()[last_ci]; // ... last chunk start long last_c1 = anyv.espc()[last_ci + 1]; // ... last chunk end Chunk[] last_cs = new Chunk[vecs.length]; // ... last chunks for (int c = 0; c < _cols.length; c++) { vecs[c] = _base.vecs()[_cols[c]]; last_cs[c] = vecs[c].chunkForChunkIdx(last_ci); } for (int i = 0; i < ix[0]._len; i++) { // select one row r = ix[0].at8(i); // next row to select if (r < 0) continue; if (r >= nrow) { for (int c = 0; c < vecs.length; c++) ncs[c].addNA(); } else { if (r < last_c0 || r >= last_c1) { last_ci = anyv.elem2ChunkIdx(r); last_c0 = anyv.espc()[last_ci]; last_c1 = anyv.espc()[last_ci + 1]; for (int c = 0; c < vecs.length; c++) last_cs[c] = vecs[c].chunkForChunkIdx(last_ci); } int ir = (int)(r - last_cs[0].start()); for (int c = 0; c < vecs.length; c++) last_cs[c].extractRows(ncs[c],ir); } } } } // Convert len rows starting at off to a 2-d ascii table @Override public String toString( ) { return ("Frame key: " + _key + "\n") + " cols: " + numCols() + "\n" + " rows: " + numRows() + "\n" + " chunks: " + (anyVec() == null ? "N/A" : anyVec().nChunks()) + "\n" + " size: " + byteSize() + "\n"; } public String toString(long off, int len) { return toTwoDimTable(off, len).toString(); } public String toString(long off, int len, boolean rollups) { return toTwoDimTable(off, len, rollups).toString(); } public TwoDimTable toTwoDimTable() { return toTwoDimTable(0,10); } public TwoDimTable toTwoDimTable(long off, int len ) { return toTwoDimTable(off,len,true); } public TwoDimTable toTwoDimTable(long off, int len, boolean rollups ) { if( off > numRows() ) off = numRows(); if( off+len > numRows() ) len = (int)(numRows()-off); String[] rowHeaders = new String[len]; int H=0; if( rollups ) { H = 5; rowHeaders = new String[len+H]; rowHeaders[0] = "min"; rowHeaders[1] = "mean"; rowHeaders[2] = "stddev"; rowHeaders[3] = "max"; rowHeaders[4] = "missing"; for( int i=0; i<len; i++ ) rowHeaders[i+H]=""+(off+i); } final int ncols = numCols(); final Vec[] vecs = vecs(); String[] coltypes = new String[ncols]; String[][] strCells = new String[len+H][ncols]; double[][] dblCells = new double[len+H][ncols]; final BufferedString tmpStr = new BufferedString(); for( int i=0; i<ncols; i++ ) { if( DKV.get(_keys[i]) == null ) { // deleted Vec in Frame coltypes[i] = "string"; for( int j=0; j<len+H; j++ ) dblCells[j][i] = TwoDimTable.emptyDouble; for( int j=0; j<len; j++ ) strCells[j+H][i] = "NO_VEC"; continue; } Vec vec = vecs[i]; if( rollups ) { dblCells[0][i] = vec.min(); dblCells[1][i] = vec.mean(); dblCells[2][i] = vec.sigma(); dblCells[3][i] = vec.max(); dblCells[4][i] = vec.naCnt(); } switch( vec.get_type() ) { case Vec.T_BAD: coltypes[i] = "string"; for( int j=0; j<len; j++ ) { strCells[j+H][i] = null; dblCells[j+H][i] = TwoDimTable.emptyDouble; } break; case Vec.T_STR : coltypes[i] = "string"; for( int j=0; j<len; j++ ) { strCells[j+H][i] = vec.isNA(off+j) ? "" : vec.atStr(tmpStr,off+j).toString(); dblCells[j+H][i] = TwoDimTable.emptyDouble; } break; case Vec.T_CAT: coltypes[i] = "string"; for( int j=0; j<len; j++ ) { strCells[j+H][i] = vec.isNA(off+j) ? "" : vec.factor(vec.at8(off+j)); dblCells[j+H][i] = TwoDimTable.emptyDouble; } break; case Vec.T_TIME: coltypes[i] = "string"; DateTimeFormatter fmt = DateTimeFormat.forPattern("yyyy-MM-dd HH:mm:ss"); for( int j=0; j<len; j++ ) { strCells[j+H][i] = vec.isNA(off+j) ? "" : fmt.print(vec.at8(off+j)); dblCells[j+H][i] = TwoDimTable.emptyDouble; } break; case Vec.T_NUM: coltypes[i] = vec.isInt() ? "long" : "double"; for( int j=0; j<len; j++ ) { dblCells[j+H][i] = vec.isNA(off+j) ? TwoDimTable.emptyDouble : vec.at(off + j); strCells[j+H][i] = null; } break; case Vec.T_UUID: throw H2O.unimpl(); default: System.err.println("bad vector type during debug print: "+vec.get_type()); throw H2O.fail(); } } return new TwoDimTable("Frame "+_key,numRows()+" rows and "+numCols()+" cols",rowHeaders,/* clone the names, the TwoDimTable will replace nulls with ""*/_names.clone(),coltypes,null, "", strCells, dblCells); } // Bulk (expensive) copy from 2nd cols into 1st cols. // Sliced by the given cols & rows private static class DeepSlice extends MRTask<DeepSlice> { final int _cols[]; final long _rows[]; final byte _isInt[]; DeepSlice( long rows[], int cols[], Vec vecs[] ) { _cols=cols; _rows=rows; _isInt = new byte[cols.length]; for( int i=0; i<cols.length; i++ ) _isInt[i] = (byte)(vecs[cols[i]].isInt() ? 1 : 0); } @Override public boolean logVerbose() { return false; } @Override public void map( Chunk chks[], NewChunk nchks[] ) { long rstart = chks[0]._start; int rlen = chks[0]._len; // Total row count int rx = 0; // Which row to in/ex-clude int rlo = 0; // Lo/Hi for this block of rows int rhi = rlen; while (true) { // Still got rows to include? if (_rows != null) { // Got a row selector? if (rx >= _rows.length) break; // All done with row selections long r = _rows[rx++];// Next row selector if (r < rstart) continue; rlo = (int) (r - rstart); rhi = rlo + 1; // Stop at the next row while (rx < _rows.length && (_rows[rx] - rstart) == rhi && rhi < rlen) { rx++; rhi++; // Grab sequential rows } } // Process this next set of rows // For all cols in the new set; BufferedString tmpStr = new BufferedString(); for (int i = 0; i < _cols.length; i++) chks[_cols[i]].extractRows(nchks[i], rlo,rhi); rlo = rhi; if (_rows == null) break; } } } /** * Create a copy of the input Frame and return that copied Frame. All Vecs in this are copied in parallel. * Caller must do the DKV.put * @param keyName Key for resulting frame. If null, no key will be given. * @return The fresh copy of fr. */ public Frame deepCopy(String keyName) { final Vec [] vecs = vecs().clone(); Key [] ks = anyVec().group().addVecs(vecs.length); Futures fs = new Futures(); for(int i = 0; i < vecs.length; ++i) DKV.put(vecs[i] = new Vec(ks[i], anyVec()._rowLayout, vecs[i].domain(),vecs()[i]._type),fs); new MRTask() { @Override public void map(Chunk[] cs) { int cidx = cs[0].cidx(); for(int i = 0; i < cs.length; ++i) DKV.put(vecs[i].chunkKey(cidx),cs[i].deepCopy(),_fs); } }.doAll(this);//.outputFrame(keyName==null?null:Key.make(keyName),this.names(),this.domains()); fs.blockForPending(); return new Frame((keyName==null?null:Key.<Frame>make(keyName)),this.names(),vecs); } /** * Last column is a bit vec indicating whether or not to take the row. */ public static class DeepSelect extends MRTask<DeepSelect> { @Override public void map( Chunk[] chks, NewChunk [] nchks ) { Chunk pred = chks[chks.length - 1]; int[] ids = pred.getIntegers(new int[pred._len],0,pred._len,0); int zeros = 0; for(int i = 0; i < ids.length; ++i) if(ids[i] == 1){ ids[i-zeros] = i; } else zeros++; ids = Arrays.copyOf(ids,ids.length-zeros); for (int c = 0; c < chks.length-1; ++c) chks[c].extractRows(nchks[c], ids); } } private String[][] domains(int [] cols){ Vec[] vecs = vecs(); String[][] res = new String[cols.length][]; for(int i = 0; i < cols.length; ++i) res[i] = vecs[cols[i]].domain(); return res; } private String [] names(int [] cols){ if(_names == null)return null; String [] res = new String[cols.length]; for(int i = 0; i < cols.length; ++i) res[i] = _names[cols[i]]; return res; } private byte[] types(int [] cols){ Vec[] vecs = vecs(); byte[] res = new byte[cols.length]; for(int i = 0; i < cols.length; ++i) res[i] = vecs[cols[i]]._type; return res; } public Vec[] makeCompatible( Frame f) {return makeCompatible(f,false);} /** Return array of Vectors if 'f' is compatible with 'this', else return a new * array of Vectors compatible with 'this' and a copy of 'f's data otherwise. Note * that this can, in the worst case, copy all of {@code this}s' data. * @return This Frame's data in an array of Vectors that is compatible with {@code f}. */ public Vec[] makeCompatible( Frame f, boolean force) { // Small data frames are always "compatible" if (anyVec() == null) // Or it is empty (no columns) return f.vecs(); // Then must be compatible Vec v1 = anyVec(); Vec v2 = f.anyVec(); if (v1 != null && v2 != null && v1.length() != v2.length()) throw new IllegalArgumentException("Can not make vectors of different length compatible!"); if (v1 == null || v2 == null || (!force && v1.isCompatibleWith(v2))) return f.vecs(); // Ok, here make some new Vecs with compatible layout Key<?> k = Key.make(); H2O.submitTask(new RebalanceDataSet(this, f, k)).join(); Frame f2 = (Frame)k.get(); DKV.remove(k); for (Vec v : f2.vecs()) Scope.track(v); return f2.vecs(); } /** * Make rows of a given frame distributed similarly to this frame. * This method is useful e.g. when we have a training frame, and we * want to adapt the testing frame so that there is proportionally * similar amount of testing data in each chunk to the amount of * training data. * The resulting frame will have the same number of chunks as this * frame. * * @param f frame that we want to re-distributed * @param newKey key for a newly created frame * @return new frame or the given frame if nothing needs to be done */ public Frame makeSimilarlyDistributed(Frame f, Key<Frame> newKey) { final Vec model = anyVec(); final Vec source = f.anyVec(); if (model == null || source == null) { // model/source frame has no columns return f; } if (model.length() == 0) { throw new IllegalStateException("Cannot make frame similarly distributed to a model frame that has no rows."); } final long[] newESPC = new long[model.espc().length]; final double ratio = (double) source.length() / model.length(); int firstNonEmpty = -1; long totalAllocated = 0; for (int i = 0; i < newESPC.length - 1; i++) { int newLen = (int) (ratio * model.chunkLen(i)); totalAllocated += newLen; newESPC[i + 1] = newLen + newESPC[i]; if (firstNonEmpty == -1 && newLen > 0) { firstNonEmpty = i; } } assert firstNonEmpty >= 0; // put rows that we didn't allocate due to rounding errors into a first non-empty chunk final int diff = (int) (source.length() - totalAllocated); assert diff >= 0; for (int i = firstNonEmpty; i < newESPC.length - 1; i++) { newESPC[i + 1] += diff; } H2O.submitTask(new RebalanceDataSet(newESPC, new Vec.VectorGroup(), f, newKey)).join(); return newKey.get(); } public static Job export(Frame fr, String path, String frameName, boolean overwrite, int nParts) { return export(fr, path, frameName, overwrite, nParts, null, new CSVStreamParams()); } public static Job export(Frame fr, String path, String frameName, boolean overwrite, int nParts, String compression, CSVStreamParams csvParms) { return export(fr, path, frameName, overwrite, nParts, false, compression, csvParms); } public static Job export(Frame fr, String path, String frameName, boolean overwrite, int nParts, boolean parallel, String compression, CSVStreamParams csvParms) { boolean forceSingle = nParts == 1; // Validate input if (H2O.getPM().isFileAccessDenied(path)) { throw new H2OFileAccessDeniedException("File " + path + " access denied"); } if (forceSingle) { boolean fileExists = H2O.getPM().exists(path); if (overwrite && fileExists) { Log.warn("File " + path + " exists, but will be overwritten!"); } else if (!overwrite && fileExists) { throw new H2OIllegalArgumentException(path, "exportFrame", "File " + path + " already exists!"); } } else { if (! H2O.getPM().isEmptyDirectoryAllNodes(path)) { throw new H2OIllegalArgumentException(path, "exportFrame", "Cannot use path " + path + " to store part files! The target needs to be either an existing empty directory or not exist yet."); } } CompressionFactory compressionFactory = compression != null ? CompressionFactory.make(compression) : null; Job job = new Job<>(fr._key, "water.fvec.Frame", "Export dataset"); FrameUtils.ExportTaskDriver t = new FrameUtils.ExportTaskDriver( fr, path, frameName, overwrite, job, nParts, parallel, compressionFactory, csvParms); return job.start(t, fr.anyVec().nChunks()); } public static Job exportParquet(Frame fr, String path, boolean overwrite, String compression, boolean writeChecksum, boolean tzAdjustFromLocal) { // Validate input if (H2O.getPM().isFileAccessDenied(path)) { throw new H2OFileAccessDeniedException("File " + path + " access denied"); } if (! H2O.getPM().isEmptyDirectoryAllNodes(path)) { throw new H2OIllegalArgumentException(path, "exportFrame", "Cannot use path " + path + " to store part files! The target needs to be either an existing empty directory or not exist yet."); } BinaryFormatExporter parquetExporter = null; ServiceLoader<BinaryFormatExporter> parquetExporters = ServiceLoader.load(BinaryFormatExporter.class); for (BinaryFormatExporter exporter : parquetExporters) { if (exporter.supports(ExportFileFormat.parquet)) { parquetExporter = exporter; break; } } if (parquetExporter == null) { Log.warn("No parquet exporter on the classpath!"); throw new RuntimeException("No parquet exporter on the classpath!"); } Job job = new Job<>(fr._key, "water.fvec.Frame", "Export dataset"); H2O.H2OCountedCompleter t = parquetExporter.export(fr, path, overwrite, compression, writeChecksum, tzAdjustFromLocal); return job.start(t, fr.anyVec().nChunks()); } /** Convert this Frame to a CSV (in an {@link InputStream}), that optionally * is compatible with R 3.1's recent change to read.csv()'s behavior. * * WARNING: Note that the end of a file is denoted by the read function * returning 0 instead of -1. * * @return An InputStream containing this Frame as a CSV */ public InputStream toCSV(CSVStreamParams parms) { return new CSVStream(this, parms); } public static class CSVStreamParams extends Iced<CSVStreamParams> { public static final char DEFAULT_SEPARATOR = ','; public static final char DEFAULT_ESCAPE = '"'; boolean _headers = true; boolean _quoteColumnNames = true; boolean _hexString = false; boolean _escapeQuotes = true; char _separator = DEFAULT_SEPARATOR; char _escapeCharacter = DEFAULT_ESCAPE; public CSVStreamParams setHeaders(boolean headers) { _headers = headers; return this; } public CSVStreamParams noHeader() { setHeaders(false); return this; } public CSVStreamParams setQuoteColumnNames(boolean quoteColumnNames) { _quoteColumnNames = quoteColumnNames; return this; } public CSVStreamParams setHexString(boolean hex_string) { _hexString = hex_string; return this; } public CSVStreamParams setSeparator(byte separator) { _separator = (char) separator; return this; } public CSVStreamParams setEscapeQuotes(boolean backslash_escape) { _escapeQuotes = backslash_escape; return this; } public CSVStreamParams setEscapeChar(char escapeChar) { _escapeCharacter = escapeChar; return this; } } public static class CSVStream extends InputStream { private static final Pattern DOUBLE_QUOTE_PATTERN = Pattern.compile("\""); private static final Set<Character> SPECIAL_CHARS = Collections.unmodifiableSet(new HashSet<>( Arrays.asList('\\', '|', '(', ')') )); private final CSVStreamParams _parms; private final Pattern escapingPattern; private final String escapeReplacement; byte[] _line; int _position; int _chkRow; Chunk[] _curChks; int _lastChkIdx; public volatile int _curChkIdx; // used only for progress reporting private transient final String[][] _escapedCategoricalVecDomains; private transient final VecEncoder[] _encoders; public CSVStream(Frame fr, CSVStreamParams parms) { this(firstChunks(fr), parms._headers ? fr.names() : null, fr.anyVec().nChunks(), parms); } public CSVStream(Chunk[] chks, String[] names, int nChunks, CSVStreamParams parms) { if (chks == null) nChunks = 0; _lastChkIdx = (chks != null) ? chks[0].cidx() + nChunks - 1 : -1; _parms = Objects.requireNonNull(parms); if (_parms._escapeCharacter != CSVStreamParams.DEFAULT_ESCAPE) { String escapeCharacterEscaped = (SPECIAL_CHARS.contains(_parms._escapeCharacter) ? "\\" : "") + _parms._escapeCharacter; escapingPattern = Pattern.compile("(\"|" + escapeCharacterEscaped + ")"); escapeReplacement = escapeCharacterEscaped + "$1"; } else { escapingPattern = DOUBLE_QUOTE_PATTERN; escapeReplacement = "\"\""; } StringBuilder sb = new StringBuilder(); if (names != null) { appendColumnName(sb, names[0]); for (int i = 1; i < names.length; i++) appendColumnName(sb.append(_parms._separator), names[i]); sb.append('\n'); } _line = StringUtils.bytesOf(sb); _chkRow = -1; // first process the header line _curChks = chks; _escapedCategoricalVecDomains = escapeCategoricalVecDomains(_curChks); if (_curChks != null) { _encoders = new VecEncoder[_curChks.length]; for (int i = 0; i < _curChks.length; i++) { Vec v = _curChks[i]._vec; if (v.isCategorical()) { _encoders[i] = VecEncoder.CAT; } else if (v.isUUID()) { _encoders[i] = VecEncoder.UUID; } else if (v.isInt()) { _encoders[i] = VecEncoder.INT; } else if (v.isString()) { _encoders[i] = VecEncoder.STRING; } else { _encoders[i] = VecEncoder.NUM; } } } else _encoders = null; } private void appendColumnName(StringBuilder sb, String name) { if (_parms._quoteColumnNames) sb.append('"'); sb.append(name); if (_parms._quoteColumnNames) sb.append('"'); } private static Chunk[] firstChunks(Frame fr) { Vec anyvec = fr.anyVec(); if (anyvec == null || anyvec.nChunks() == 0 || anyvec.length() == 0) { return null; } Chunk[] chks = new Chunk[fr.vecs().length]; for (int i = 0; i < fr.vecs().length; i++) { chks[i] = fr.vec(i).chunkForRow(0); } return chks; } /** * Escapes categorical levels of vectors and puts them in a map of escaped categorical levels. * Only the domains with at least one level with an escaped quote are saved. If a domain does not need * any escaping, it is considered better practice to reach to the `vec.domain()` method itself and not duplicate entries * in memory here. * * @param chunks * @return A 2D array of String[][]. Elements can be null of give domain does not need escaping. */ private String[][] escapeCategoricalVecDomains(final Chunk[] chunks) { if(chunks == null) return null; final String[][] localEscapedCategoricalVecDomains = new String[chunks.length][]; for (int i = 0; i < chunks.length; i++) { final Vec vec = chunks[i].vec(); if (!vec.isCategorical()) continue; final String[] originalDomain = vec.domain(); final String[] escapedDomain = new String[originalDomain.length]; for (int level = 0; level < originalDomain.length; level++) { escapedDomain[level] = '"' + escapeQuotesForCsv(originalDomain[level]) + '"'; } localEscapedCategoricalVecDomains[i] = escapedDomain; } return localEscapedCategoricalVecDomains; } public int getCurrentRowSize() { int av = available(); assert av > 0; return _line.length; } enum VecEncoder {CAT, UUID, INT, STRING, NUM}; byte[] getBytesForRow() { StringBuilder sb = new StringBuilder(); final BufferedString unescapedTempStr = new BufferedString(); for (int i = 0; i < _curChks.length; i++) { if (i > 0) sb.append(_parms._separator); if (!_curChks[i].isNA(_chkRow)) { switch (_encoders[i]) { case NUM: // R 3.1 unfortunately changed the behavior of read.csv(). // (Really type.convert()). // // Numeric values with too much precision now trigger a type conversion in R 3.1 into a factor. // // See these discussions: // https://bugs.r-project.org/bugzilla/show_bug.cgi?id=15751 // https://stat.ethz.ch/pipermail/r-devel/2014-April/068778.html // http://stackoverflow.com/questions/23072988/preserve-old-pre-3-1-0-type-convert-behavior double d = _curChks[i].atd(_chkRow); String s = _parms._hexString ? Double.toHexString(d) : RyuDouble.doubleToString(d); sb.append(s); break; case CAT: final String escapedCat = _escapedCategoricalVecDomains[i][(int) _curChks[i].at8(_chkRow)]; sb.append(escapedCat); break; case INT: sb.append(_curChks[i].at8(_chkRow)); break; case STRING: final String escapedString = escapeQuotesForCsv(_curChks[i].atStr(unescapedTempStr, _chkRow).toString()); sb.append('"').append(escapedString).append('"'); break; case UUID: sb.append(PrettyPrint.UUID(_curChks[i].at16l(_chkRow), _curChks[i].at16h(_chkRow))); break; default: throw new IllegalStateException("Unknown encoder " + _encoders[i]); } } } sb.append('\n'); return StringUtils.bytesOf(sb); } /** * Escapes double-quotes (ASCII 34) in a String. * * @param unescapedString An unescaped {@link String} to escape * @return String with escaped double-quotes, if found. */ private String escapeQuotesForCsv(final String unescapedString) { if (!_parms._escapeQuotes) return unescapedString; return escapingPattern.matcher(unescapedString).replaceAll(escapeReplacement); } @Override public int available() { // Case 1: There is more data left to read from the current line. if (_position != _line.length) { return _line.length - _position; } // Case 2: There are no chunks to work with (eg. the whole Frame was empty). if (_curChks == null) { return -1; } _chkRow++; Chunk anyChunk = _curChks[0]; // Case 3: Out of data. if (anyChunk._start + _chkRow >= anyChunk._vec.length()) { return -1; } // Case 4: Out of data in the current chunks => fast-forward to the next set of non-empty chunks. if (_chkRow == anyChunk.len()) { _curChkIdx = anyChunk._vec.elem2ChunkIdx(anyChunk._start + _chkRow); // skips empty chunks // Case 4: Processed all requested chunks. if (_curChkIdx > _lastChkIdx) { return -1; } // fetch the next non-empty chunks Chunk[] newChks = new Chunk[_curChks.length]; for (int i = 0; i < _curChks.length; i++) { newChks[i] = _curChks[i]._vec.chunkForChunkIdx(_curChkIdx); // flush the remote chunk Key oldKey = _curChks[i]._vec.chunkKey(_curChks[i]._cidx); if (! oldKey.home()) { H2O.raw_remove(oldKey); } } _curChks = newChks; _chkRow = 0; } // Case 5: Return data for the current row. _line = getBytesForRow(); _position = 0; return _line.length; } @Override public void close() throws IOException { super.close(); _line = null; } @Override public int read() throws IOException { return available() == 0 ? -1 : _line[_position++]; } @Override public int read(byte[] b, int off, int len) throws IOException { int n = available(); if(n > 0) { n = Math.min(n, len); System.arraycopy(_line, _position, b, off, n); _position += n; } return n; } } @Override public Class<KeyV3.FrameKeyV3> makeSchema() { return KeyV3.FrameKeyV3.class; } /** Sort rows of a frame, using the set of columns as keys. User can specify sorting direction for each sorting * column in a integer array. For example, if we want to sort columns 0, 1, 2 and want to sort 0 in ascending * direction, 1 and 2 in descending direction for frame fr, the call to make is fr.sort(new int[]{0,1,2}, * new int[]{1, -1, -1}. * * @return Copy of frame, sorted */ public Frame sort( int[] cols ) { return Merge.sort(this,cols); } public Frame sort(int[] cols, int[] ascending) { return Merge.sort(this, cols, ascending); } /** * A structure for fast lookup in the set of frame's vectors. * Purpose of this class is to avoid multiple O(n) searches in {@link Frame}'s vectors. * * @return An instance of {@link FrameVecRegistry} */ public FrameVecRegistry frameVecRegistry() { return new FrameVecRegistry(); } /** * Returns the original frame with specific column converted to categorical */ public Frame toCategoricalCol(int columIdx){ write_lock(); replace(columIdx, vec(columIdx).toCategoricalVec()).remove(); // Update frame in DKV update(); unlock(); return this; } /** * Returns the original frame with specific column converted to categorical */ public Frame toCategoricalCol(String column){ return toCategoricalCol(find(column)); } public class FrameVecRegistry { private LinkedHashMap<String, Vec> vecMap; private FrameVecRegistry() { vecMap = new LinkedHashMap<>(_vecs.length); for (int i = 0; i < _vecs.length; i++) { vecMap.put(_names[i], _vecs[i]); } } /** * Finds a Vec by column name * * @param colName Column name to search for, case-sensitive * @return An instance of {@link Vec}, if found. Otherwise null. */ public Vec findByColName(final String colName) { return vecMap.get(colName); } } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/FrameCreator.java
package water.fvec; import hex.CreateFrame; import jsr166y.CountedCompleter; import water.*; import water.util.ArrayUtils; import water.util.FrameUtils; import java.util.Arrays; import java.util.Random; import java.util.UUID; import static water.fvec.Vec.makeCon; import water.util.RandomUtils; /** * Helper to make up a Frame from scratch, with random content */ public class FrameCreator extends H2O.H2OCountedCompleter { transient Vec _v; final private CreateFrame _createFrame; private int[] _cat_cols; private int[] _int_cols; private int[] _real_cols; private int[] _bin_cols; private int[] _time_cols; private int[] _string_cols; private String[][] _domain; private Frame _out; public FrameCreator(CreateFrame createFrame) { _createFrame = createFrame; int[] idx = _createFrame.has_response ? ArrayUtils.seq(1, _createFrame.cols + 1) : ArrayUtils.seq(0, _createFrame.cols); int[] shuffled_idx = new int[idx.length]; ArrayUtils.shuffleArray(idx, idx.length, shuffled_idx, _createFrame.seed_for_column_types, 0); // Sometimes the client requests, say, 0.3 categorical columns. By the time this number arrives here, it becomes // something like 0.299999999997. If we just multiply by the number of columns (say 10000) and take integer part, // we'd have 2999 columns only -- not what the client expects. This is why we add 0.1 to each count before taking // the floor part. int catcols = (int)(_createFrame.categorical_fraction * _createFrame.cols + 0.1); int intcols = (int)(_createFrame.integer_fraction * _createFrame.cols + 0.1); int bincols = (int)(_createFrame.binary_fraction * _createFrame.cols + 0.1); int timecols = (int)(_createFrame.time_fraction * _createFrame.cols + 0.1); int stringcols = (int)(_createFrame.string_fraction * _createFrame.cols + 0.1); int realcols = _createFrame.cols - catcols - intcols - bincols - timecols - stringcols; // At this point we might accidentally allocated too many columns. In such a case, adjust their counts. if (realcols < 0 && catcols > 0) { catcols--; realcols++; } if (realcols < 0 && intcols > 0) { intcols--; realcols++; } if (realcols < 0 && bincols > 0) { bincols--; realcols++; } if (realcols < 0 && timecols > 0) { timecols--; realcols++; } if (realcols < 0 && stringcols > 0) { stringcols--; realcols++; } assert(catcols >= 0); assert(intcols >= 0); assert(bincols >= 0); assert(realcols >= 0); assert(timecols >= 0); assert(stringcols >= 0); _cat_cols = Arrays.copyOfRange(shuffled_idx, 0, catcols); _int_cols = Arrays.copyOfRange(shuffled_idx, catcols, catcols+intcols); _real_cols = Arrays.copyOfRange(shuffled_idx, catcols+intcols, catcols+intcols+realcols); _bin_cols = Arrays.copyOfRange(shuffled_idx, catcols+intcols+realcols, catcols+intcols+realcols+bincols); _time_cols = Arrays.copyOfRange(shuffled_idx, catcols+intcols+realcols+bincols, catcols+intcols+realcols+bincols+timecols); _string_cols = Arrays.copyOfRange(shuffled_idx, catcols+intcols+realcols+bincols+timecols, catcols+intcols+realcols+bincols+timecols+stringcols); // create domains for categorical variables _domain = new String[_createFrame.cols + (_createFrame.has_response ? 1 : 0)][]; if(createFrame.randomize) { if(_createFrame.has_response) { assert (_createFrame.response_factors >= 1); _domain[0] = _createFrame.response_factors == 1 ? null : new String[_createFrame.response_factors]; if (_domain[0] != null) { for (int i = 0; i < _domain[0].length; ++i) { _domain[0][i] = Integer.toString(i); } } } for (int c : _cat_cols) { _domain[c] = new String[_createFrame.factors]; for (int i = 0; i < _createFrame.factors; ++i) { _domain[c][i] = "c" + c + ".l" + i; } } } // All columns together fill one chunk final int rows_per_chunk = FileVec.calcOptimalChunkSize( (int)((float)(catcols+intcols)*_createFrame.rows*4 //4 bytes for categoricals and integers +(float)bincols *_createFrame.rows*1*_createFrame.binary_ones_fraction //sparse uses a fraction of one byte (or even less) +(float)(realcols+timecols+stringcols) *_createFrame.rows*8), //8 bytes for real and time (long) values _createFrame.cols, _createFrame.cols*4, H2ORuntime.availableProcessors(), H2O.getCloudSize(), false, true); _v = makeCon(_createFrame.value, _createFrame.rows, (int)Math.ceil(Math.log1p(rows_per_chunk)),false); } public int nChunks() { return _v.nChunks(); } @Override public void compute2() { int totcols = _createFrame.cols + (_createFrame.has_response ? 1 : 0); Vec[] vecs = new Vec[totcols]; if(_createFrame.randomize) { byte[] types = new byte[vecs.length]; for (int i : _cat_cols) types[i] = Vec.T_CAT; for (int i : _bin_cols) types[i] = Vec.T_NUM; for (int i : _int_cols) types[i] = Vec.T_NUM; for (int i : _real_cols) types[i] = Vec.T_NUM; for (int i : _time_cols) types[i] = Vec.T_TIME; for (int i : _string_cols) types[i] = Vec.T_STR; if (_createFrame.has_response) { types[0] = _createFrame.response_factors == 1 ? Vec.T_NUM : Vec.T_CAT; } vecs = _v.makeZeros(totcols, _domain, types); } else { for (int i = 0; i < vecs.length; ++i) vecs[i] = _v.makeCon(_createFrame.value); } _v.remove(); _v=null; String[] names = new String[vecs.length]; if(_createFrame.has_response) { names[0] = "response"; for (int i = 1; i < vecs.length; i++) names[i] = "C" + i; } else { for (int i = 0; i < vecs.length; i++) names[i] = "C" + (i+1); } _out = new Frame(_createFrame._job._result, names, vecs); assert _out.numRows() == _createFrame.rows; assert _out.numCols() == totcols; _out.delete_and_lock(_createFrame._job._key); // fill with random values new FrameRandomizer(_createFrame, _cat_cols, _int_cols, _real_cols, _bin_cols, _time_cols, _string_cols).doAll(_out); //overwrite a fraction with N/A FrameUtils.MissingInserter mi = new FrameUtils.MissingInserter(_createFrame._job._result, _createFrame.seed, _createFrame.missing_fraction); mi.execImpl().get(); tryComplete(); } @Override public void onCompletion(CountedCompleter caller){ _out.update(_createFrame._job._key); _out.unlock(_createFrame._job._key); } private static class FrameRandomizer extends MRTask<FrameRandomizer> { final private CreateFrame _createFrame; final private int[] _cat_cols; final private int[] _int_cols; final private int[] _real_cols; final private int[] _bin_cols; final private int[] _time_cols; final private int[] _string_cols; public FrameRandomizer(CreateFrame createFrame, int[] cat_cols, int[] int_cols, int[] real_cols, int[] bin_cols, int[] time_cols, int[] string_cols){ _createFrame = createFrame; _cat_cols = cat_cols; _int_cols = int_cols; _real_cols = real_cols; _bin_cols = bin_cols; _time_cols = time_cols; _string_cols = string_cols; } //row+col-dependent RNG for reproducibility with different number of VMs, chunks, etc. void setSeed(Random rng, int col, long row) { rng.setSeed(_createFrame.seed + _createFrame.cols * row + col); rng.setSeed(rng.nextLong()); } @Override public void map (Chunk[]cs){ Job<Frame> job = _createFrame._job; if (job.stop_requested()) return; if (!_createFrame.randomize) return; final Random rng = RandomUtils.getRNG(new Random().nextLong()); // response if(_createFrame.has_response) { for (int r = 0; r < cs[0]._len; r++) { setSeed(rng, 0, cs[0]._start + r); if (_createFrame.response_factors > 1) cs[0].set(r, (int) (rng.nextDouble() * _createFrame.response_factors)); //classification else if (_createFrame.positive_response) cs[0].set(r, _createFrame.real_range * rng.nextDouble()); //regression with positive response else cs[0].set(r, _createFrame.real_range * (1 - 2 * rng.nextDouble())); //regression } } job.update(1); for (int c : _cat_cols) { for (int r = 0; r < cs[c]._len; r++) { setSeed(rng, c, cs[c]._start + r); cs[c].set(r, (int)(rng.nextDouble() * _createFrame.factors)); } } job.update(1); for (int c : _int_cols) { for (int r = 0; r < cs[c]._len; r++) { setSeed(rng, c, cs[c]._start + r); cs[c].set(r, -_createFrame.integer_range + (long)(rng.nextDouble()*(2*_createFrame.integer_range+1))); } } job.update(1); for (int c : _real_cols) { for (int r = 0; r < cs[c]._len; r++) { setSeed(rng, c, cs[c]._start + r); cs[c].set(r, _createFrame.real_range * (1 - 2 * rng.nextDouble())); } } job.update(1); for (int c : _bin_cols) { for (int r = 0; r < cs[c]._len; r++) { setSeed(rng, c, cs[c]._start + r); cs[c].set(r, rng.nextFloat() > _createFrame.binary_ones_fraction ? 0 : 1); } } job.update(1); for (int c : _time_cols) { for (int r = 0; r < cs[c]._len; r++) { setSeed(rng, c, cs[c]._start + r); cs[c].set(r, Math.abs(rng.nextLong() % (50L*365*24*3600*1000))); //make a random moment in time between 1970 and 2020 } } job.update(1); byte[] by = new byte[8]; for (int c : _string_cols) { for (int r = 0; r < cs[c]._len; r++) { setSeed(rng, c, cs[c]._start + r); for (int i=0;i<by.length;++i) by[i] = (byte)(65+rng.nextInt(25)); cs[c].set(r, new String(by)); } } job.update(1); } } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/GcsFileVec.java
package water.fvec; import water.DKV; import water.Futures; import water.Key; import water.Value; public class GcsFileVec extends FileVec { private GcsFileVec(Key key, long len) { super(key, len, Value.GCS); } public static Key make(String path, long size) { Futures fs = new Futures(); Key<Frame> key = make(path, size, fs); fs.blockForPending(); return key; } public static Key<Frame> make(String path, long size, Futures fs) { Key<Frame> frameKey = Key.make(path); Key<Vec> vecKey = Vec.newKey(frameKey); new Frame(frameKey).delete_and_lock(); // Insert the top-level FileVec key into the store Vec vec = new GcsFileVec(vecKey, size); DKV.put(vecKey, vec, fs); Frame frame = new Frame(frameKey, new String[]{path}, new Vec[]{vec}); frame.update(); frame.unlock(); return frameKey; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/HDFSFileVec.java
package water.fvec; import water.*; /** * Vec representation of file stored on HDFS. */ public final class HDFSFileVec extends FileVec { private HDFSFileVec(Key key, long len) { super(key, len, Value.HDFS); } public static Key make(String path, long size) { Futures fs = new Futures(); Key key = make(path, size, fs); fs.blockForPending(); return key; } public static Key make(String path, long size, Futures fs) { Key k = Key.make(path); Key k2 = Vec.newKey(k); new Frame(k).delete_and_lock(); // Insert the top-level FileVec key into the store Vec v = new HDFSFileVec(k2,size); DKV.put(k2, v, fs); Frame fr = new Frame(k,new String[]{path},new Vec[]{v}); fr.update(); fr.unlock(); return k; } @Override public byte[] getFirstBytes() { return FileVecUtils.getFirstBytes(this); } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/HTTPFileVec.java
package water.fvec; import water.*; /** * FileVec backed by an HTTP/HTTPS data source */ public class HTTPFileVec extends FileVec { private HTTPFileVec(Key key, long len) { super(key, len, Value.HTTP); } public static Key make(String path, long size) { Futures fs = new Futures(); Key key = make(path, size, fs); fs.blockForPending(); return key; } public static Key make(String path, long size, Futures fs) { Key k = Key.make(path); Key k2 = Vec.newKey(k); new Frame(k).delete_and_lock(); // Insert the top-level FileVec key into the store Vec v = new HTTPFileVec(k2, size); DKV.put(k2, v, fs); Frame fr = new Frame(k, new String[]{path}, new Vec[]{v}); fr.update(); fr.unlock(); return k; } }
0
java-sources/ai/h2o/h2o-core/3.46.0.7/water
java-sources/ai/h2o/h2o-core/3.46.0.7/water/fvec/InteractionWrappedVec.java
package water.fvec; import water.*; import water.util.ArrayUtils; import water.util.IcedHashMap; import water.util.IcedLong; import java.util.Arrays; import java.util.Collections; import java.util.HashSet; import java.util.Map; /** * * This class represents an interaction between two Vec instances. * * Another virtual Vec (akin to TransformWrappedVec) used to represent a * never-materialized interaction between two columns. * * There are 3 types of interactions to consider: Num-Num, Num-Enum, and Enum-Enums * Creation of these Vec instances is cheap except for the Enum-Enum case (since it is not * known apriori the co-occurrence of enums between any two categorical columns). So in * this specific case, an MRTask is done to collect the domain. * * * @author spencer */ public class InteractionWrappedVec extends WrappedVec { private final Key<Vec> _masterVecKey1; private final Key<Vec> _masterVecKey2; private transient Vec _masterVec1; private transient Vec _masterVec2; private String[] _v1Domain; private String[] _v2Domain; public boolean _useAllFactorLevels; public boolean _skipMissing; public boolean _standardize; private long[] _bins; private String[] _missingDomains; public transient GetMeanTask t; private String[] _v1Enums; // only interact these enums from vec 1 private String[] _v2Enums; // only interact these enums from vec 2 public InteractionWrappedVec(Key<Vec> key, int rowLayout, String[] vec1DomainLimit, String[] vec2DomainLimit, boolean useAllFactorLevels, boolean skipMissing, boolean standardize, Key<Vec> masterVecKey1, Key<Vec> masterVecKey2) { super(key, rowLayout, null); _masterVecKey1=masterVecKey1; _masterVecKey2=masterVecKey2; _v1Enums=vec1DomainLimit; _v2Enums=vec2DomainLimit; _masterVec1=_masterVecKey1.get(); _masterVec2=_masterVecKey2.get(); _useAllFactorLevels=useAllFactorLevels; _skipMissing=skipMissing; setupDomain(_standardize=standardize); // performs MRTask if both vecs are categorical!! DKV.put(this); if (t != null) t.doAll(this); } public String[] v1Domain() { return _v1Enums==null?_v1Domain:_v1Enums; } public String[] v2Domain() { return _v2Enums==null?_v2Domain:_v2Enums; } @Override public String[] domain() { // always returns the "correct" domains, so accidental mixup of domain vs domains is ok String[] res1 = v1Domain(); String[] res2 = v2Domain(); return res1 == null? res2 : res2 == null? res1 : super.domain(); } public Vec v1() { return _masterVec1==null?(_masterVec1=_masterVecKey1.get()):_masterVec1; } public Vec v2() { return _masterVec2==null?(_masterVec2=_masterVecKey2.get()):_masterVec2; } /** * Obtain the length of the expanded (i.e. one-hot expanded) interaction column. */ public int expandedLength() { if (isNumericInteraction()) return 1; // 2 numeric columns -> 1 column else if( isCategorical() ) return domain().length; // 2 cat -> domains (limited) length else if( _v1Domain!=null ) return _v1Enums==null?_v1Domain.length - (_useAllFactorLevels?0:1):_v1Enums.length-(_useAllFactorLevels?0:1); else return _v2Enums==null?_v2Domain.length - (_useAllFactorLevels?0:1):_v2Enums.length - (_useAllFactorLevels?0:1); } public double[] getMeans() { if( null!=_v1Domain && null!=_v2Domain ) { double[] res = new double[domain().length]; Arrays.fill(res,Double.NaN); return res; } else if( null==_v1Domain && null==_v2Domain ) return new double[]{super.mean()}; return new GetMeanTask(v1Domain()==null?v2Domain().length:v1Domain().length).doAll(this)._d; } public double getSub(int i) { if (t == null) return mean(); return t._d[i]; } public double getMul(int i) { double sigma = (t == null)? sigma() : t._sigma[i]; return sigma == 0? 1.0 : 1.0/sigma; } public double getSigma(int i) { return (t == null)? sigma() : t._sigma[i]; } private static class GetMeanTask extends MRTask<GetMeanTask> { private double[] _d; // means, NA skipped private double[] _sigma; // sds, NA skipped private long _rows; private final int _len; GetMeanTask(int len) { _len=len; } @Override public void map(Chunk c) { _d = new double[_len]; _sigma = new double[_len]; InteractionWrappedChunk cc = (InteractionWrappedChunk)c; Chunk lC = cc._c[0]; Chunk rC = cc._c[1]; // get the "left" chk and the "rite" chk if( cc._c2IsCat ) { lC=rC; rC=cc._c[0]; } // left is always cat long rows=0; for(int rid=0;rid<c._len;++rid) { if( lC.isNA(rid) || rC.isNA(rid) ) continue; // skipmissing int idx = (int)lC.at8(rid); rows++; for(int i=0;i<_d.length;++i) { double x = i==idx?rC.atd(rid):0; double delta = x - _d[i]; _d[i] += delta / rows; _sigma[i] += delta * (x - _d[i]); } } _rows=rows; } @Override public void reduce(GetMeanTask t) { if (_rows == 0) { _d = t._d; _sigma = t._sigma; } else if(t._rows != 0){ for(int i=0;i<_d.length;++i) { double delta = _d[i] - t._d[i]; _d[i] = (_d[i]* _rows + t._d[i] * t._rows) / (_rows + t._rows); _sigma[i] += t._sigma[i] + delta * delta * _rows * t._rows / (_rows + t._rows); } } _rows += t._rows; } @Override public void postGlobal() { for(int i=0;i<_sigma.length;++i ) _sigma[i] = Math.sqrt(_sigma[i]/(_rows-1)); } } public boolean isNumericInteraction() { return null==v1Domain() && null==v2Domain(); } @Override public double mean() { if (null==t && isNumericInteraction()) return super.mean(); return 0; } @Override public double sigma() { if (null==t && isNumericInteraction()) return super.sigma(); return 1; } @Override public int mode() { if( !isCategorical() ) { if (v1Domain() != null) { assert v2Domain() == null; // Only one of them can have a domain (it would be categorical otherwise) return _masterVec1.mode(); } if (v2Domain() != null) { assert v1Domain() == null; // Only one of them can have a domain (it would be categorical otherwise) return _masterVec2.mode(); } // Numerical interactions don't have a mode throw H2O.unimpl(); } return ArrayUtils.maxIndex(_bins); } public long[] getBins() { return _bins; } public String[] missingDomains() { return _missingDomains; } private void setupDomain(boolean standardize) { if( _masterVec1.isCategorical() || _masterVec2.isCategorical() ) { _v1Domain = _masterVec1.domain(); _v2Domain = _masterVec2.domain(); if( _v1Domain!=null && _v2Domain!=null ) { CombineDomainTask t =new CombineDomainTask(_v1Domain, _v2Domain,_v1Enums,_v2Enums, _useAllFactorLevels,_skipMissing).doAll(_masterVec1, _masterVec2); setDomain(t._dom); _bins=t._bins; _type = Vec.T_CAT; // vec is T_NUM up to this point _missingDomains=t._missingDom; } else t = standardize?new GetMeanTask(v1Domain()==null?v2Domain().length:v1Domain().length):null; } if( null==_v1Domain && null==_v2Domain ) _useAllFactorLevels=true; // just makes life easier to have this when the vec is categorical } private static class CombineDomainTask extends MRTask<CombineDomainTask> { private String[] _dom; // out, sorted (uses Arrays.sort) private long[] _bins; // out, sorted according to _dom private String[] _missingDom; // out, the missing levels due to !_useAllLvls private final String _left[]; // in private final String _rite[]; // in private final String _leftLimit[]; // in private final String _riteLimit[]; // in private final boolean _useAllLvls; // in private final boolean _skipMissing; // in private IcedHashMap<String, IcedLong> _perChkMap; private IcedHashMap<String, String> _perChkMapMissing; // skipped cats CombineDomainTask(String[] left, String[] rite, String[] leftLimit, String[] riteLimit, boolean useAllLvls, boolean skipMissing) { _left = left; _rite = rite; _leftLimit = leftLimit; _riteLimit = riteLimit; _useAllLvls = useAllLvls; _skipMissing = skipMissing; } @Override public void map(Chunk[] c) { _perChkMap = new IcedHashMap<>(); if( !_useAllLvls ) _perChkMapMissing = new IcedHashMap<>(); Chunk left = c[0]; Chunk rite = c[1]; String k; HashSet<String> A = _leftLimit == null ? null : new HashSet<String>(); HashSet<String> B = _riteLimit == null ? null : new HashSet<String>(); if (A != null) Collections.addAll(A, _leftLimit); if (B != null) Collections.addAll(B, _riteLimit); int lval,rval; String l,r; boolean leftIsNA, riteIsNA; for (int i = 0; i < left._len; ++i) if( (!((leftIsNA=left.isNA(i)) | (riteIsNA=rite.isNA(i)))) ) { lval = (int)left.at8(i); rval = (int)rite.at8(i); if( !_useAllLvls && ( 0==lval || 0==rval )) { // deal with all illegal levels here when _useAllLvls==False _perChkMapMissing.putIfAbsent(_left[lval] + "_" + _rite[rval],""); continue; } l = _left[lval]; r = _rite[rval]; if (A != null && !A.contains(l)) continue; if (B != null && !B.contains(r)) continue; if( null!=_perChkMap.putIfAbsent((k = l + "_" + r), new IcedLong(1)) ) _perChkMap.get(k)._val++; } else if( !_skipMissing ) { if( !(leftIsNA && riteIsNA) ) { // not both missing if( leftIsNA ) { r = _rite[rval=(int)rite.at8(i)]; if( !_useAllLvls && 0==rval ) { _perChkMapMissing.putIfAbsent("NA_" + _rite[rval],""); continue; } if( B!=null && !B.contains(r) ) continue; if( null!=_perChkMap.putIfAbsent((k="NA_"+r), new IcedLong(1)) ) _perChkMap.get(k)._val++; } else { l = _left[lval=(int)left.at8(i)]; if( !_useAllLvls && 0==lval ) { _perChkMapMissing.putIfAbsent(_left[lval] + "_NA",""); continue; } if( null!=A && !A.contains(l) ) continue; if( null!=_perChkMap.putIfAbsent((k=l+"_NA"), new IcedLong(1)) ) _perChkMap.get(k)._val++; } } } } @Override public void reduce(CombineDomainTask t) { for (Map.Entry<String, IcedLong> e : t._perChkMap.entrySet()) { IcedLong i = _perChkMap.get(e.getKey()); if (i != null) i._val += e.getValue()._val; else _perChkMap.put(e.getKey(), e.getValue()); } t._perChkMap = null; if(_perChkMapMissing==null && t._perChkMapMissing!=null ) { _perChkMapMissing=new IcedHashMap<>(); _perChkMapMissing.putAll(t._perChkMapMissing); } else if( _perChkMapMissing!=null && t._perChkMapMissing!=null ) { for (String s: t._perChkMapMissing.keySet()) _perChkMapMissing.putIfAbsent(s,""); } t._perChkMapMissing=null; } @Override public void postGlobal() { Arrays.sort(_dom = _perChkMap.keySet().toArray(new String[_perChkMap.size()])); int idx = 0; _bins = new long[_perChkMap.size()]; for(String s:_dom) _bins[idx++] = _perChkMap.get(s)._val; if( _missingDom!=null ) Arrays.sort(_missingDom = _perChkMapMissing.keySet().toArray(new String[_perChkMapMissing.size()])); } } @Override public Chunk chunkForChunkIdx(int cidx) { Chunk[] cs = new Chunk[2]; cs[0] = (_masterVec1!=null?_masterVec1: (_masterVec1=_masterVecKey1.get())).chunkForChunkIdx(cidx); cs[1] = (_masterVec2!=null?_masterVec2: (_masterVec2=_masterVecKey2.get())).chunkForChunkIdx(cidx); return new InteractionWrappedChunk(this, cs); } @Override public Vec doCopy() { InteractionWrappedVec v = new InteractionWrappedVec(group().addVec(), _rowLayout,_v1Enums,_v2Enums, _useAllFactorLevels, _skipMissing, _standardize, _masterVecKey1, _masterVecKey2); if( null!=domain() ) v.setDomain(domain()); if( null!=_v1Domain ) v._v1Domain=_v1Domain.clone(); if( null!=_v2Domain ) v._v2Domain=_v2Domain.clone(); return v; } @Override protected AutoBuffer writeAll_impl(AutoBuffer ab) { ab.putAStr(_v1Domain); ab.putAStr(_v2Domain); ab.putZ(_useAllFactorLevels); ab.putZ(_skipMissing); ab.putZ(_standardize); ab.putAStr(_missingDomains); ab.putAStr(_v1Enums); ab.putAStr(_v2Enums); return super.writeAll_impl(ab); } @Override protected Keyed readAll_impl(AutoBuffer ab, Futures fs) { _v1Domain=ab.getAStr(); _v2Domain=ab.getAStr(); _useAllFactorLevels=ab.getZ(); _skipMissing=ab.getZ(); _standardize=ab.getZ(); _missingDomains=ab.getAStr(); _v1Enums=ab.getAStr(); _v2Enums=ab.getAStr(); return super.readAll_impl(ab,fs); } public static class InteractionWrappedChunk extends Chunk { public final transient Chunk[] _c; public final boolean _c1IsCat; // left chunk is categorical public final boolean _c2IsCat; // rite chunk is categorical public final boolean _isCat; // this vec is categorical InteractionWrappedChunk(InteractionWrappedVec transformWrappedVec, Chunk[] c) { // set all the chunk fields _c = c; set_len(_c[0]._len); _start = _c[0]._start; _vec = transformWrappedVec; _cidx = _c[0]._cidx; _c1IsCat=_c[0]._vec.isCategorical(); _c2IsCat=_c[1]._vec.isCategorical(); _isCat = _vec.isCategorical(); } @Override public double atd_impl(int idx) { if( _isCat ) if( isNA_impl(idx) ) return Double.NaN; if (_isCat) { int val = Arrays.binarySearch(_vec.domain(), getKey(idx)); return ((val < 0)?-1:val); } else return ( _c1IsCat?1: (_c[0].atd(idx))) * ( _c2IsCat?1: (_c[1].atd(idx)) ); } @Override public long at8_impl(int idx) { if (_isCat) { long val = Arrays.binarySearch(_vec.domain(), getKey(idx)); // can give bad value like -3 return ((val < 0)?-1:val); // illegal domain should always return -1 } else return ( _c1IsCat?1:_c[0].at8(idx) ) * ( _c2IsCat?1:_c[1].at8(idx) ); } private String getKey(int idx) { return _c[0]._vec.domain()[(int)_c[0].at8(idx)] + "_" + _c[1]._vec.domain()[(int)_c[1].at8(idx)]; } @Override public boolean isNA_impl(int idx) { return _c[0].isNA(idx) || _c[1].isNA(idx); } // Returns true if the masterVec is missing, false otherwise @Override public boolean set_impl(int idx, long l) { return false; } @Override public boolean set_impl(int idx, double d) { return false; } @Override public boolean set_impl(int idx, float f) { return false; } @Override public boolean setNA_impl(int idx) { return false; } @Override public ChunkVisitor processRows(ChunkVisitor nc, int from, int to){ for(int i = from; i < to; i++) nc.addValue(atd(i)); return nc; } @Override public ChunkVisitor processRows(ChunkVisitor nc, int... rows){ for(int i:rows) nc.addValue(atd(i)); return nc; } @Override protected final void initFromBytes () { throw water.H2O.fail(); } } }