Update vision and audio models

app.py

@@ -11,6 +11,8 @@ from transformers.models.auto.configuration_auto import CONFIG_MAPPING_NAMES
 from transformers.models.auto.modeling_auto import (
     MODEL_FOR_CTC_MAPPING_NAMES,
     MODEL_FOR_AUDIO_CLASSIFICATION_MAPPING_NAMES,
+    MODEL_FOR_AUDIO_FRAME_CLASSIFICATION_MAPPING_NAMES,
+    MODEL_FOR_AUDIO_XVECTOR_MAPPING_NAMES,
     MODEL_FOR_SPEECH_SEQ_2_SEQ_MAPPING_NAMES,
     MODEL_FOR_VISION_2_SEQ_MAPPING_NAMES,
     MODEL_FOR_INSTANCE_SEGMENTATION_MAPPING_NAMES,
@@ -18,11 +20,23 @@ from transformers.models.auto.modeling_auto import (
     MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING_NAMES,
     MODEL_FOR_CAUSAL_IMAGE_MODELING_MAPPING_NAMES,
     MODEL_FOR_MASKED_IMAGE_MODELING_MAPPING_NAMES,
+    MODEL_FOR_TABLE_QUESTION_ANSWERING_MAPPING_NAMES,
+    MODEL_FOR_VISUAL_QUESTION_ANSWERING_MAPPING_NAMES,
+    MODEL_FOR_DOCUMENT_QUESTION_ANSWERING_MAPPING_NAMES,
+    MODEL_FOR_BACKBONE_MAPPING_NAMES,
+    _MODEL_FOR_ZERO_SHOT_IMAGE_CLASSIFICATION_MAPPING_NAMES,
 )
 
-audio_models = list(MODEL_FOR_CTC_MAPPING_NAMES.keys()) + list(MODEL_FOR_AUDIO_CLASSIFICATION_MAPPING_NAMES.keys()) + list(MODEL_FOR_SPEECH_SEQ_2_SEQ_MAPPING_NAMES.keys())
+audio_models = list(MODEL_FOR_CTC_MAPPING_NAMES.keys()) + list(MODEL_FOR_AUDIO_CLASSIFICATION_MAPPING_NAMES.keys()) + \
+               list(MODEL_FOR_SPEECH_SEQ_2_SEQ_MAPPING_NAMES.keys()) + list(MODEL_FOR_AUDIO_FRAME_CLASSIFICATION_MAPPING_NAMES.keys()) + \
+               list(MODEL_FOR_AUDIO_XVECTOR_MAPPING_NAMES.keys())
 
-vision_models = ["clip"] + list(MODEL_FOR_VISION_2_SEQ_MAPPING_NAMES.keys()) + list(MODEL_FOR_INSTANCE_SEGMENTATION_MAPPING_NAMES.keys()) + list(MODEL_FOR_SEMANTIC_SEGMENTATION_MAPPING_NAMES.keys()) + list(MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING_NAMES.keys()) + list(MODEL_FOR_CAUSAL_IMAGE_MODELING_MAPPING_NAMES.keys()) + list(MODEL_FOR_MASKED_IMAGE_MODELING_MAPPING_NAMES.keys())
+vision_models = list(MODEL_FOR_VISION_2_SEQ_MAPPING_NAMES.keys()) + list(MODEL_FOR_INSTANCE_SEGMENTATION_MAPPING_NAMES.keys()) + \
+                list(MODEL_FOR_SEMANTIC_SEGMENTATION_MAPPING_NAMES.keys()) + list(MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING_NAMES.keys()) + \
+                list(MODEL_FOR_CAUSAL_IMAGE_MODELING_MAPPING_NAMES.keys()) + list(MODEL_FOR_MASKED_IMAGE_MODELING_MAPPING_NAMES.keys()) + \
+                list(MODEL_FOR_TABLE_QUESTION_ANSWERING_MAPPING_NAMES.keys()) + list(MODEL_FOR_VISUAL_QUESTION_ANSWERING_MAPPING_NAMES.keys()) + \
+                list(MODEL_FOR_DOCUMENT_QUESTION_ANSWERING_MAPPING_NAMES.keys()) + list(MODEL_FOR_BACKBONE_MAPPING_NAMES.keys()) + \
+                list(_MODEL_FOR_ZERO_SHOT_IMAGE_CLASSIFICATION_MAPPING_NAMES.keys())
 
 today = datetime.date.today()
 year, week, _ = today.isocalendar()
@@ -30,6 +44,7 @@ year, week, _ = today.isocalendar()
 DATASET_REPO_URL = (
     "https://huggingface.co/datasets/huggingface/transformers-stats-space-data"
 )
+
 DATA_FILENAME = f"data_{week}_{year}.csv"
 DATA_FILE = os.path.join("data", DATA_FILENAME)
 
@@ -65,9 +80,12 @@ def retrieve_model_stats():
             [m.downloads for m in models if hasattr(m, "downloads")]
         )
         if len(models) > 0:
-            model_stats["download_per_model"] = round(
-                model_stats["num_downloads"] / len(models), 2
+            model_stats["download_per_model"] = int(
+                model_stats["num_downloads"] / len(models)
             )
+        else:
+            model_stats["download_per_model"] = model_stats["num_downloads"]
+
         total_downloads += model_stats["num_downloads"]
 
         # save in overall dict
@@ -106,7 +124,6 @@ if not os.path.isfile(DATA_FILE):
 with open(DATA_FILE, "r") as f:
     dataframe = pd.read_csv(DATA_FILE)
 
-dataframe[dataframe["modality"] == "audio"]
 int_downloads = np.array(
     [int(x.replace(",", "")) for x in dataframe["num_downloads"].values]
 )
@@ -198,7 +215,7 @@ st.title("All stats last 30 days")
 st.table(dataframe)
 
 st.title("Vision stats last 30 days")
-st.table(dataframe[dataframe["modality"] == "vision"])
+st.table(dataframe[dataframe["modality"] == "vision"].drop("modality", axis=1))
 
 st.title("Audio stats last 30 days")
-st.table(dataframe[dataframe["modality"] == "audio"])
+st.table(dataframe[dataframe["modality"] == "audio"].drop("modality", axis=1))

requirements.txt

@@ -1,2 +1,2 @@
-transformers>=4.24.0
+git+https://github.com/huggingface/transformers.git
 huggingface_hub