Upload folder using huggingface_hub

config.json

@@ -1,13 +1,7 @@
 {
-  "_name_or_path": "models/checkpoints/dictabert_large/P1_add_weights/best_model",
-  "architectures": [
-    "BertForTokenClassification"
-  ],
-  "attention_probs_dropout_prob": 0.1,
-  "classifier_dropout": null,
-  "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.1,
-  "hidden_size": 1024,
+  "model_type": "bert-crf",
+  "base_model_name": "dicta-il/dictabert",
+  "num_labels": 9,
   "id2label": {
     "0": "O",
     "1": "B-SUBSTITUTION",
@@ -19,30 +13,37 @@
     "7": "B-ADDITION",
     "8": "I-ADDITION"
   },
-  "initializer_range": 0.02,
-  "intermediate_size": 4096,
   "label2id": {
-    "B-ADDITION": 7,
-    "B-QUANTITY": 3,
+    "O": 0,
     "B-SUBSTITUTION": 1,
-    "B-TECHNIQUE": 5,
-    "I-ADDITION": 8,
-    "I-QUANTITY": 4,
     "I-SUBSTITUTION": 2,
+    "B-QUANTITY": 3,
+    "I-QUANTITY": 4,
+    "B-TECHNIQUE": 5,
     "I-TECHNIQUE": 6,
-    "O": 0
+    "B-ADDITION": 7,
+    "I-ADDITION": 8
   },
-  "layer_norm_eps": 1e-12,
-  "max_position_embeddings": 512,
-  "model_type": "bert",
-  "newmodern": true,
-  "num_attention_heads": 16,
-  "num_hidden_layers": 24,
-  "pad_token_id": 0,
-  "position_embedding_type": "absolute",
-  "torch_dtype": "float32",
-  "transformers_version": "4.48.0",
-  "type_vocab_size": 2,
-  "use_cache": true,
-  "vocab_size": 128000
-}
+  "dropout_rate": 0.1,
+  "focal_gamma": null,
+  "class_weights": null,
+  "architecture": "BertCRFModel",
+  "train_args": {
+    "model": "dicta-il/dictabert",
+    "dropout": 0.1,
+    "data_dir": "data/processed_v2",
+    "train_file": "train_merged.jsonl",
+    "max_len": 128,
+    "class_weights_file": "data/processed_v2/stats_merged.json",
+    "output_dir": "models/checkpoints/dictabert_crf/P1_crf_weighted",
+    "epochs": 10,
+    "batch_size": 16,
+    "lr": 2e-05,
+    "patience": 3,
+    "seed": 42,
+    "neg_ratio": null,
+    "wandb": false,
+    "wandb_project": "recipe-mod-extraction",
+    "wandb_run": null
+  }
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75b0cafe3b9a5f56088b14d7a2a0bb3d9d7bf27dbef2353587f4c3b6074088a7
+size 737490298

tokenizer.json

@@ -75,14 +75,14 @@
         "pattern": {
           "String": "<foreign>"
         },
-        "content": "[UNK]"
+        "content": "UNK"
       },
       {
         "type": "Replace",
         "pattern": {
-          "Regex": "[^֐-׿\u0000-‌-‿₠-₿∀-⋿⅐-↋ff-ﭏ]+"
+          "Regex": "[^֐-׿\u0000-‌-‿₠-₿∀-⋿⅐-↋ff-ﭏ]+"
         },
-        "content": "[UNK]"
+        "content": "UNK"
       }
     ]
   },