Upload 6 files

.gitattributes

@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text
+processed_output.jsonl filter=lfs diff=lfs merge=lfs -text

gliner_config.json

@@ -1,10 +1,10 @@
 {
+  "bos_token_id": 2,
   "class_token_index": 256000,
   "decoder_mode": null,
-  "dropout": 0.4,
+  "dropout": 0.3,
   "embed_ent_token": true,
   "encoder_config": {
-    "_attn_implementation_autoset": false,
     "_name_or_path": "jhu-clsp/mmBERT-base",
     "add_cross_attention": false,
     "architectures": [
@@ -27,6 +27,7 @@
     "deterministic_flash_attn": false,
     "diversity_penalty": 0.0,
     "do_sample": false,
+    "dtype": "float32",
     "early_stopping": false,
     "embedding_dropout": 0.0,
     "encoder_no_repeat_ngram_size": 0,
@@ -80,7 +81,6 @@
     "prefix": null,
     "problem_type": null,
     "pruned_heads": {},
-    "reference_compile": null,
     "remove_invalid_values": false,
     "repad_logits_with_grad": false,
     "repetition_penalty": 1.0,
@@ -98,13 +98,13 @@
     "tokenizer_class": null,
     "top_k": 50,
     "top_p": 1.0,
-    "torch_dtype": "float32",
     "torchscript": false,
     "typical_p": 1.0,
     "use_bfloat16": false,
     "vocab_size": 256002
   },
   "ent_token": "<<ENT>>",
+  "eos_token_id": 1,
   "eval_every": 500,
   "fine_tune": true,
   "freeze_token_rep": false,
@@ -117,25 +117,26 @@
   "labels_decoder_config": null,
   "labels_encoder": null,
   "labels_encoder_config": null,
-  "log_dir": "models/mmbert_gliner",
+  "log_dir": "models/",
   "loss_alpha": 0.75,
   "loss_gamma": 0,
   "loss_prob_margin": 0,
   "loss_reduction": "sum",
-  "lr_encoder": "5e-6",
+  "lr_encoder": "2e-5",
   "lr_others": "3e-5",
   "masking": "none",
   "max_grad_norm": 10.0,
-  "max_len": 8192,
+  "max_len": 2048,
   "max_neg_type_ratio": 1,
-  "max_types": 100,
+  "max_types": 200,
   "max_width": 12,
   "model_name": "jhu-clsp/mmBERT-base",
   "model_type": "gliner",
-  "name": "mmBERT GLiNER NER",
+  "name": "span level gliner",
   "negatives": 1.0,
   "num_post_fusion_layers": 1,
-  "num_steps": 150,
+  "num_steps": 100000,
+  "pad_token_id": 0,
   "post_fusion_schema": null,
   "prev_path": null,
   "random_drop": true,
@@ -147,10 +148,10 @@
   "size_sup": -1,
   "span_mode": "markerV0",
   "subtoken_pooling": "first",
-  "train_batch_size": 4,
-  "train_data": "data/sample_data.json",
-  "transformers_version": "4.48.1",
-  "val_data_dir": "none",
+  "train_batch_size": 16,
+  "train_data": "../dataset_gen/dataset_generated/processed_output.jsonl",
+  "transformers_version": "4.57.1",
+  "val_data": "../dataset_gen/dataset_generated/test_it.jsonl",
   "vocab_size": 256002,
   "warmup_ratio": 0.05,
   "weight_decay_encoder": 0.1,

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f2b1a08c8ba688f939cadfd317ae894cbca471c1e54cf9b14157244c318732d
-size 34363556
+oid sha256:4f65190517e4633347d84e0225b8c5c66a1a62321107ebc73c09d5cefa2b5cd5
+size 34363810