LASER / src /vine_hf /vine_pipeline.py
moqingyan123
final fixes
888f9e4
raw
history blame
30.4 kB
import torch
import numpy as np
import cv2
import os
from typing import Dict, List, Tuple, Optional, Any, Union
from transformers import Pipeline
import tempfile
import uuid
from .vine_config import VineConfig
from .vine_model import VineModel
from .vis_utils import render_dino_frames, render_sam_frames, render_vine_frame_sets
from laser.loading import load_video
from laser.preprocess.mask_generation_grounding_dino import generate_masks_grounding_dino
class VinePipeline(Pipeline):
"""
Pipeline for VINE model that handles end-to-end video understanding.
This pipeline takes a video file or frames, along with segmentation method
and keyword lists, and returns probability distributions over the keywords.
Segmentation Model Configuration:
The pipeline requires SAM2 and GroundingDINO models for mask generation.
You can configure custom paths via constructor kwargs:
- sam_config_path: Path to SAM2 config (e.g., "configs/sam2.1/sam2.1_hiera_b+.yaml")
- sam_checkpoint_path: Path to SAM2 checkpoint (e.g., "checkpoints/sam2.1_hiera_base_plus.pt")
- gd_config_path: Path to GroundingDINO config (e.g., "groundingdino/config/GroundingDINO_SwinT_OGC.py")
- gd_checkpoint_path: Path to GroundingDINO checkpoint (e.g., "checkpoints/groundingdino_swint_ogc.pth")
Old:
- SAM2: ~/research/sam2/ or /home/asethi04/LASER_NEW/LASER/sam2/
- GroundingDINO: /home/asethi04/LASER_NEW/LASER/GroundingDINO/
Alternative: Use set_segmentation_models() to provide pre-initialized model instances.
"""
def __init__(
self,
sam_config_path: Optional[str] = None,
sam_checkpoint_path: Optional[str] = None,
gd_config_path: Optional[str] = None,
gd_checkpoint_path: Optional[str] = None,
**kwargs
):
self.grounding_model = None
self.sam_predictor = None
self.mask_generator = None
self.sam_config_path = sam_config_path
self.sam_checkpoint_path = sam_checkpoint_path
self.gd_config_path = gd_config_path
self.gd_checkpoint_path = gd_checkpoint_path
super().__init__(**kwargs)
# Set default parameters from config
self.segmentation_method = getattr(self.model.config, 'segmentation_method', 'grounding_dino_sam2')
self.box_threshold = getattr(self.model.config, 'box_threshold', 0.35)
self.text_threshold = getattr(self.model.config, 'text_threshold', 0.25)
self.target_fps = getattr(self.model.config, 'target_fps', 1)
self.visualize = getattr(self.model.config, 'visualize', False)
self.visualization_dir = getattr(self.model.config, 'visualization_dir', None)
self.debug_visualizations = getattr(self.model.config, 'debug_visualizations', False)
self._device = getattr(self.model.config, '_device')
if kwargs.get("device") is not None:
self._device = kwargs.get("device")
def set_segmentation_models(
self,
*,
sam_predictor=None,
mask_generator=None,
grounding_model=None
):
"""
Set pre-initialized segmentation models, bypassing automatic initialization/current_values
Args:
sam_predictor: Pre-built SAM2 video predictor
mask_generator: Pre-built SAM2 automatic mask generator
grounding_model: Pre-built GroundingDINO model
"""
if sam_predictor is not None:
self.sam_predictor = sam_predictor
if mask_generator is not None:
self.mask_generator = mask_generator
if grounding_model is not None:
self.grounding_model = grounding_model
def _sanitize_parameters(self, **kwargs):
"""Sanitize parameters for different pipeline stages."""
preprocess_kwargs = {}
forward_kwargs = {}
postprocess_kwargs = {}
# Preprocess parameters
if "segmentation_method" in kwargs:
preprocess_kwargs["segmentation_method"] = kwargs["segmentation_method"]
if "target_fps" in kwargs:
preprocess_kwargs["target_fps"] = kwargs["target_fps"]
if "box_threshold" in kwargs:
preprocess_kwargs["box_threshold"] = kwargs["box_threshold"]
if "text_threshold" in kwargs:
preprocess_kwargs["text_threshold"] = kwargs["text_threshold"]
if "categorical_keywords" in kwargs:
preprocess_kwargs["categorical_keywords"] = kwargs["categorical_keywords"]
# Forward parameters
if "categorical_keywords" in kwargs:
forward_kwargs["categorical_keywords"] = kwargs["categorical_keywords"]
if "unary_keywords" in kwargs:
forward_kwargs["unary_keywords"] = kwargs["unary_keywords"]
if "binary_keywords" in kwargs:
forward_kwargs["binary_keywords"] = kwargs["binary_keywords"]
if "object_pairs" in kwargs:
forward_kwargs["object_pairs"] = kwargs["object_pairs"]
if "return_flattened_segments" in kwargs:
forward_kwargs["return_flattened_segments"] = kwargs["return_flattened_segments"]
if "return_valid_pairs" in kwargs:
forward_kwargs["return_valid_pairs"] = kwargs["return_valid_pairs"]
if "interested_object_pairs" in kwargs:
forward_kwargs["interested_object_pairs"] = kwargs["interested_object_pairs"]
if "debug_visualizations" in kwargs:
forward_kwargs["debug_visualizations"] = kwargs["debug_visualizations"]
postprocess_kwargs["debug_visualizations"] = kwargs["debug_visualizations"]
# Postprocess parameters
if "return_top_k" in kwargs:
postprocess_kwargs["return_top_k"] = kwargs["return_top_k"]
if "self.visualize" in kwargs:
postprocess_kwargs["self.visualize"] = kwargs["self.visualize"]
return preprocess_kwargs, forward_kwargs, postprocess_kwargs
def preprocess(
self,
video_input: Union[str, np.ndarray, torch.Tensor],
segmentation_method: str = None,
target_fps: int = None,
box_threshold: float = None,
text_threshold: float = None,
categorical_keywords: List[str] = None,
**kwargs
) -> Dict[str, Any]:
"""
Preprocess video input and generate masks.
Args:
video_input: Path to video file, or video tensor/array
segmentation_method: "sam2" or "grounding_dino_sam2"
target_fps: Target FPS for video processing
box_threshold: Box threshold for Grounding DINO
text_threshold: Text threshold for Grounding DINO
categorical_keywords: Keywords for Grounding DINO segmentation
Returns:
Dict containing video frames, masks, and bboxes
"""
# Use defaults from config if not provided
if segmentation_method is None:
segmentation_method = self.segmentation_method
if target_fps is None:
target_fps = self.target_fps
if box_threshold is None:
box_threshold = self.box_threshold
if text_threshold is None:
text_threshold = self.text_threshold
if categorical_keywords is None:
categorical_keywords = ["object"] # Default generic category
if isinstance(video_input, str):
# Video file path
video_tensor = load_video(video_input, target_fps=target_fps)
if isinstance(video_tensor, list):
video_tensor = np.array(video_tensor)
elif isinstance(video_tensor, torch.Tensor):
video_tensor = video_tensor.cpu().numpy()
elif isinstance(video_input, (np.ndarray, torch.Tensor)):
# Video tensor/array
if isinstance(video_input, torch.Tensor):
video_tensor = video_input.numpy()
else:
video_tensor = video_input
else:
raise ValueError(f"Unsupported video input type: {type(video_input)}")
# Ensure video tensor is numpy array
if not isinstance(video_tensor, np.ndarray):
video_tensor = np.array(video_tensor)
# Ensure video tensor is in correct format
if len(video_tensor.shape) != 4:
raise ValueError(f"Expected video tensor shape (frames, height, width, channels), got {video_tensor.shape}")
# Generate masks and bboxes based on segmentation method
visualization_data: Dict[str, Any] = {}
print(f"Segmentation method: {segmentation_method}")
if segmentation_method == "sam2":
masks, bboxes, vis_data = self._generate_sam2_masks(video_tensor)
elif segmentation_method == "grounding_dino_sam2":
masks, bboxes, vis_data = self._generate_grounding_dino_sam2_masks(
video_tensor, categorical_keywords, box_threshold, text_threshold, video_input
)
else:
raise ValueError(f"Unsupported segmentation method: {segmentation_method}")
if vis_data:
visualization_data.update(vis_data)
visualization_data.setdefault("sam_masks", masks)
return {
"video_frames": torch.tensor(video_tensor),
"masks": masks,
"bboxes": bboxes,
"num_frames": len(video_tensor),
"visualization_data": visualization_data,
}
def _generate_sam2_masks(self, video_tensor: np.ndarray) -> Tuple[Dict, Dict, Dict[str, Any]]:
"""Generate masks using SAM2 automatic mask generation."""
# Initialize SAM2 models if not already done
print("Generating SAM2 masks...")
if self.mask_generator is None:
self._initialize_segmentation_models()
if self.mask_generator is None:
raise ValueError("SAM2 mask generator not available")
masks: Dict[int, Dict[int, torch.Tensor]] = {}
bboxes: Dict[int, Dict[int, List[int]]] = {}
for frame_id, frame in enumerate(video_tensor):
if isinstance(frame, np.ndarray) and frame.dtype != np.uint8:
frame = (frame * 255).astype(np.uint8) if frame.max() <= 1 else frame.astype(np.uint8)
height, width, _ = frame.shape
frame_masks = self.mask_generator.generate(frame)
masks[frame_id] = {}
bboxes[frame_id] = {}
for obj_id, mask_data in enumerate(frame_masks):
mask = mask_data["segmentation"]
if isinstance(mask, np.ndarray):
mask = torch.from_numpy(mask)
if len(mask.shape) == 2:
mask = mask.unsqueeze(-1)
elif len(mask.shape) == 3 and mask.shape[0] == 1:
mask = mask.permute(1, 2, 0)
wrapped_id = obj_id + 1
masks[frame_id][wrapped_id] = mask
mask_np = mask.squeeze().numpy() if isinstance(mask, torch.Tensor) else mask.squeeze()
coords = np.where(mask_np > 0)
if len(coords[0]) > 0:
y1, y2 = coords[0].min(), coords[0].max()
x1, x2 = coords[1].min(), coords[1].max()
bboxes[frame_id][wrapped_id] = [x1, y1, x2, y2]
return masks, bboxes, {"sam_masks": masks}
def _generate_grounding_dino_sam2_masks(
self,
video_tensor: np.ndarray,
categorical_keywords: List[str],
box_threshold: float,
text_threshold: float,
video_path: str,
) -> Tuple[Dict, Dict, Dict[str, Any]]:
"""Generate masks using Grounding DINO + SAM2."""
# Initialize models if not already done
print("Generating Grounding DINO + SAM2 masks...")
if self.grounding_model is None or self.sam_predictor is None:
self._initialize_segmentation_models()
if self.grounding_model is None or self.sam_predictor is None:
raise ValueError("GroundingDINO or SAM2 models not available")
temp_video_path = None
if video_path is None or not isinstance(video_path, str):
temp_video_path = self._create_temp_video(video_tensor)
video_path = temp_video_path
CHUNK = 5
classes_ls = [categorical_keywords[i:i + CHUNK] for i in range(0, len(categorical_keywords), CHUNK)]
video_segments, oid_class_pred, _ = generate_masks_grounding_dino(
self.grounding_model,
box_threshold,
text_threshold,
self.sam_predictor,
self.mask_generator,
video_tensor,
video_path,
"temp_video",
out_dir=tempfile.gettempdir(),
classes_ls=classes_ls,
target_fps=self.target_fps,
visualize=self.debug_visualizations,
frames=None,
max_prop_time=10
)
masks: Dict[int, Dict[int, torch.Tensor]] = {}
bboxes: Dict[int, Dict[int, List[int]]] = {}
for frame_id, frame_masks in video_segments.items():
masks[frame_id] = {}
bboxes[frame_id] = {}
for obj_id, mask in frame_masks.items():
if not isinstance(mask, torch.Tensor):
mask = torch.tensor(mask)
masks[frame_id][obj_id] = mask
mask_np = mask.numpy()
if mask_np.ndim == 3 and mask_np.shape[0] == 1:
mask_np = np.squeeze(mask_np, axis=0)
coords = np.where(mask_np > 0)
if len(coords[0]) > 0:
y1, y2 = coords[0].min(), coords[0].max()
x1, x2 = coords[1].min(), coords[1].max()
bboxes[frame_id][obj_id] = [x1, y1, x2, y2]
if temp_video_path and os.path.exists(temp_video_path):
os.remove(temp_video_path)
vis_data: Dict[str, Any] = {
"sam_masks": masks,
"dino_labels": oid_class_pred,
}
return masks, bboxes, vis_data
def _initialize_segmentation_models(self):
"""Initialize segmentation models based on the requested method and configured paths."""
if (self.sam_predictor is None or self.mask_generator is None):
self._initialize_sam2_models()
if self.grounding_model is None:
self._initialize_grounding_dino_model()
def _initialize_sam2_models(self):
"""Initialize SAM2 video predictor and mask generator."""
try:
from sam2.build_sam import build_sam2_video_predictor, build_sam2
from sam2.automatic_mask_generator import SAM2AutomaticMaskGenerator
except ImportError as e:
print(f"Warning: Could not import SAM2: {e}")
return
# Resolve SAM2 paths
config_path, checkpoint_path = self._resolve_sam2_paths()
# Validate paths if custom ones were provided
if self.sam_config_path is not None and not os.path.exists(config_path):
raise ValueError(f"SAM2 config path not found: {config_path}")
if self.sam_checkpoint_path is not None and not os.path.exists(checkpoint_path):
raise ValueError(f"SAM2 checkpoint path not found: {checkpoint_path}")
# Only proceed if we have valid paths
if not os.path.exists(checkpoint_path):
print(f"Warning: SAM2 checkpoint not found at {checkpoint_path}")
print("SAM2 functionality will be unavailable")
return
try:
device = self._device
print(type(device))
# Video predictor
self.sam_predictor = build_sam2_video_predictor(
config_path, checkpoint_path, device=device
)
# Mask generator
sam2_model = build_sam2(config_path, checkpoint_path, device=device, apply_postprocessing=False)
self.mask_generator = SAM2AutomaticMaskGenerator(
model=sam2_model,
points_per_side=32,
points_per_batch=32,
pred_iou_thresh=0.7,
stability_score_thresh=0.8,
crop_n_layers=2,
box_nms_thresh=0.6,
crop_n_points_downscale_factor=2,
min_mask_region_area=100,
use_m2m=True,
)
print("✓ SAM2 models initialized successfully")
except Exception as e:
raise ValueError(f"Failed to initialize SAM2 with custom paths: {e}")
def _initialize_grounding_dino_model(self):
"""Initialize GroundingDINO model."""
try:
from groundingdino.util.inference import Model as gd_Model
except ImportError as e:
print(f"Warning: Could not import GroundingDINO: {e}")
return
# Resolve GroundingDINO paths
config_path, checkpoint_path = self._resolve_grounding_dino_paths()
# Validate paths if custom ones were provided
if self.gd_config_path is not None and not os.path.exists(config_path):
raise ValueError(f"GroundingDINO config path not found: {config_path}")
if self.gd_checkpoint_path is not None and not os.path.exists(checkpoint_path):
raise ValueError(f"GroundingDINO checkpoint path not found: {checkpoint_path}")
# Only proceed if we have valid paths
if not (os.path.exists(config_path) and os.path.exists(checkpoint_path)):
print(f"Warning: GroundingDINO models not found at {config_path} / {checkpoint_path}")
print("GroundingDINO functionality will be unavailable")
return
try:
device = self._device
print(type(device))
self.grounding_model = gd_Model(
model_config_path=config_path,
model_checkpoint_path=checkpoint_path,
device=device
)
print("✓ GroundingDINO model initialized successfully")
except Exception as e:
raise ValueError(f"Failed to initialize GroundingDINO with custom paths: {e}")
def _resolve_sam2_paths(self):
"""Resolve SAM2 config and checkpoint paths."""
# Use custom paths if provided
if self.sam_config_path and self.sam_checkpoint_path:
return self.sam_config_path, self.sam_checkpoint_path
def _resolve_grounding_dino_paths(self):
"""Resolve GroundingDINO config and checkpoint paths."""
# Use custom paths if provided
if self.gd_config_path and self.gd_checkpoint_path:
return self.gd_config_path, self.gd_checkpoint_path
def _prepare_visualization_dir(self, name: str, enabled: bool) -> Optional[str]:
"""
Ensure a directory exists for visualization artifacts and return it.
If visualization is disabled, returns None.
"""
if not enabled:
return None
if self.visualization_dir:
target_dir = os.path.join(self.visualization_dir, name) if name else self.visualization_dir
os.makedirs(target_dir, exist_ok=True)
return target_dir
return tempfile.mkdtemp(prefix=f"vine_{name}_")
def _create_temp_video(self, video_tensor: np.ndarray, base_dir: Optional[str] = None, prefix: str = "temp_video") -> str:
"""Create a temporary video file from video tensor."""
if base_dir is None:
base_dir = tempfile.mkdtemp(prefix=f"vine_{prefix}_")
else:
os.makedirs(base_dir, exist_ok=True)
file_name = f"{prefix}_{uuid.uuid4().hex}.mp4"
temp_path = os.path.join(base_dir, file_name)
# Use OpenCV to write video
height, width = video_tensor.shape[1:3]
fourcc = cv2.VideoWriter_fourcc(*'mp4v')
out = cv2.VideoWriter(temp_path, fourcc, self.target_fps, (width, height))
for frame in video_tensor:
# Convert RGB to BGR for OpenCV
if len(frame.shape) == 3 and frame.shape[2] == 3:
frame_bgr = cv2.cvtColor(frame, cv2.COLOR_RGB2BGR)
else:
frame_bgr = frame
out.write(frame_bgr.astype(np.uint8))
out.release()
return temp_path
def _forward(self, model_inputs: Dict[str, Any], **forward_kwargs) -> Dict[str, Any]:
"""Forward pass through the model."""
outputs = self.model.predict(
video_frames=model_inputs["video_frames"],
masks=model_inputs["masks"],
bboxes=model_inputs["bboxes"],
**forward_kwargs
)
outputs.setdefault("video_frames", model_inputs.get("video_frames"))
outputs.setdefault("bboxes", model_inputs.get("bboxes"))
outputs.setdefault("masks", model_inputs.get("masks"))
outputs.setdefault("visualization_data", model_inputs.get("visualization_data"))
return outputs
def postprocess(
self,
model_outputs: Dict[str, Any],
return_top_k: int = 3,
visualize: Optional[bool] = None,
**kwargs
) -> Dict[str, Any]:
"""
Postprocess model outputs into user-friendly format.
Args:
model_outputs: Raw model outputs
return_top_k: Number of top predictions to return
self.visualize: Whether to include visualization data
Returns:
Formatted results
"""
results = {
"categorical_predictions": model_outputs.get("categorical_predictions", {}),
"unary_predictions": model_outputs.get("unary_predictions", {}),
"binary_predictions": model_outputs.get("binary_predictions", {}),
"confidence_scores": model_outputs.get("confidence_scores", {}),
"summary": self._generate_summary(model_outputs)
}
if "flattened_segments" in model_outputs:
results["flattened_segments"] = model_outputs["flattened_segments"]
if "valid_pairs" in model_outputs:
results["valid_pairs"] = model_outputs["valid_pairs"]
if "valid_pairs_metadata" in model_outputs:
results["valid_pairs_metadata"] = model_outputs["valid_pairs_metadata"]
if "visualization_data" in model_outputs:
results["visualization_data"] = model_outputs["visualization_data"]
if self.visualize and "video_frames" in model_outputs and "bboxes" in model_outputs:
frames_tensor = model_outputs["video_frames"]
if isinstance(frames_tensor, torch.Tensor):
frames_np = frames_tensor.detach().cpu().numpy()
else:
frames_np = np.asarray(frames_tensor)
if frames_np.dtype != np.uint8:
if np.issubdtype(frames_np.dtype, np.floating):
max_val = frames_np.max() if frames_np.size else 0.0
scale = 255.0 if max_val <= 1.0 else 1.0
frames_np = (frames_np * scale).clip(0, 255).astype(np.uint8)
else:
frames_np = frames_np.clip(0, 255).astype(np.uint8)
cat_label_lookup: Dict[int, Tuple[str, float]] = {}
for obj_id, preds in model_outputs.get("categorical_predictions", {}).items():
if preds:
prob, label = preds[0]
cat_label_lookup[obj_id] = (label, prob)
unary_preds = model_outputs.get("unary_predictions", {})
unary_lookup: Dict[int, Dict[int, List[Tuple[float, str]]]] = {}
for (frame_id, obj_id), preds in unary_preds.items():
if preds:
unary_lookup.setdefault(frame_id, {})[obj_id] = preds
binary_preds = model_outputs.get("binary_predictions", {})
binary_lookup: Dict[int, List[Tuple[Tuple[int, int], List[Tuple[float, str]]]]] = {}
for (frame_id, obj_pair), preds in binary_preds.items():
if preds:
binary_lookup.setdefault(frame_id, []).append((obj_pair, preds))
bboxes = model_outputs["bboxes"]
visualization_data = model_outputs.get("visualization_data", {})
visualizations: Dict[str, Dict[str, Any]] = {}
debug_visualizations = kwargs.get("debug_visualizations")
if debug_visualizations is None:
debug_visualizations = self.debug_visualizations
vine_frame_sets = render_vine_frame_sets(
frames_np,
bboxes,
cat_label_lookup,
unary_lookup,
binary_lookup,
visualization_data.get("sam_masks"),
)
vine_visuals: Dict[str, Dict[str, Any]] = {}
final_frames = vine_frame_sets.get("all", [])
if final_frames:
final_entry: Dict[str, Any] = {"frames": final_frames, "video_path": None}
final_dir = self._prepare_visualization_dir("all", enabled=self.visualize)
final_entry["video_path"] = self._create_temp_video(
np.stack(final_frames, axis=0),
base_dir=final_dir,
prefix="all_visualization"
)
vine_visuals["all"] = final_entry
if debug_visualizations:
sam_masks = visualization_data.get("sam_masks")
if sam_masks:
sam_frames = render_sam_frames(frames_np, sam_masks, visualization_data.get("dino_labels"))
sam_entry = {"frames": sam_frames, "video_path": None}
if sam_frames:
sam_dir = self._prepare_visualization_dir("sam", enabled=self.visualize)
sam_entry["video_path"] = self._create_temp_video(
np.stack(sam_frames, axis=0),
base_dir=sam_dir,
prefix="sam_visualization"
)
visualizations["sam"] = sam_entry
dino_labels = visualization_data.get("dino_labels")
if dino_labels:
dino_frames = render_dino_frames(frames_np, bboxes, dino_labels)
dino_entry = {"frames": dino_frames, "video_path": None}
if dino_frames:
dino_dir = self._prepare_visualization_dir("dino", enabled=self.visualize)
dino_entry["video_path"] = self._create_temp_video(
np.stack(dino_frames, axis=0),
base_dir=dino_dir,
prefix="dino_visualization"
)
visualizations["dino"] = dino_entry
for name in ("object", "unary", "binary"):
frames_list = vine_frame_sets.get(name, [])
entry: Dict[str, Any] = {"frames": frames_list, "video_path": None}
if frames_list:
vine_dir = self._prepare_visualization_dir(name, enabled=self.visualize)
entry["video_path"] = self._create_temp_video(
np.stack(frames_list, axis=0),
base_dir=vine_dir,
prefix=f"{name}_visualization"
)
vine_visuals[name] = entry
if vine_visuals:
visualizations["vine"] = vine_visuals
if visualizations:
results["visualizations"] = visualizations
return results
def _generate_summary(self, model_outputs: Dict[str, Any]) -> Dict[str, Any]:
"""Generate a summary of the predictions."""
categorical_preds = model_outputs.get("categorical_predictions", {})
unary_preds = model_outputs.get("unary_predictions", {})
binary_preds = model_outputs.get("binary_predictions", {})
summary = {
"num_objects_detected": len(categorical_preds),
"num_unary_predictions": len(unary_preds),
"num_binary_predictions": len(binary_preds),
"top_categories": [],
"top_actions": [],
"top_relations": []
}
# Extract top categories
all_categories = []
for obj_preds in categorical_preds.values():
if obj_preds:
all_categories.extend(obj_preds)
if all_categories:
sorted_categories = sorted(all_categories, reverse=True)
summary["top_categories"] = [(cat, prob) for prob, cat in sorted_categories[:3]]
# Extract top actions
all_actions = []
for action_preds in unary_preds.values():
if action_preds:
all_actions.extend(action_preds)
if all_actions:
sorted_actions = sorted(all_actions, reverse=True)
summary["top_actions"] = [(act, prob) for prob, act in sorted_actions[:3]]
# Extract top relations
all_relations = []
for rel_preds in binary_preds.values():
if rel_preds:
all_relations.extend(rel_preds)
if all_relations:
sorted_relations = sorted(all_relations, reverse=True)
summary["top_relations"] = [(rel, prob) for prob, rel in sorted_relations[:3]]
return summary