Updated for cnlp v0.6.0

added_tokens.json

@@ -1 +1,10 @@
-{"<cr>": 50271, "</e>": 50266, "</a2>": 50270, "</a1>": 50268, "<neg>": 50272, "<a1>": 50267, "<e>": 50265, "<a2>": 50269}
+{
+  "</a1>": 28898,
+  "</a2>": 28900,
+  "</e>": 28896,
+  "<a1>": 28897,
+  "<a2>": 28899,
+  "<cr>": 28901,
+  "<e>": 28895,
+  "<neg>": 28902
+}

config.json

@@ -2,15 +2,16 @@
   "architectures": [
     "CnlpModelForClassification"
   ],
+  "cnlpt_version": "0.6.0",
   "encoder_config": {
-    "_name_or_path": "roberta-base",
+    "_name_or_path": "microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract",
     "add_cross_attention": false,
     "architectures": [
-      "RobertaForMaskedLM"
+      "BertForMaskedLM"
     ],
     "attention_probs_dropout_prob": 0.1,
     "bad_words_ids": null,
-    "bos_token_id": 0,
+    "bos_token_id": null,
     "chunk_size_feed_forward": 0,
     "classifier_dropout": null,
     "cross_attention_hidden_size": null,
@@ -19,7 +20,8 @@
     "do_sample": false,
     "early_stopping": false,
     "encoder_no_repeat_ngram_size": 0,
-    "eos_token_id": 2,
+    "eos_token_id": null,
+    "exponential_decay_length_penalty": null,
     "finetuning_task": null,
     "forced_bos_token_id": null,
     "forced_eos_token_id": null,
@@ -38,12 +40,12 @@
       "LABEL_0": 0,
       "LABEL_1": 1
     },
-    "layer_norm_eps": 1e-05,
+    "layer_norm_eps": 1e-12,
     "length_penalty": 1.0,
     "max_length": 20,
-    "max_position_embeddings": 514,
+    "max_position_embeddings": 512,
     "min_length": 0,
-    "model_type": "roberta",
+    "model_type": "bert",
     "no_repeat_ngram_size": 0,
     "num_attention_heads": 12,
     "num_beam_groups": 1,
@@ -53,7 +55,7 @@
     "output_attentions": false,
     "output_hidden_states": false,
     "output_scores": false,
-    "pad_token_id": 1,
+    "pad_token_id": 0,
     "position_embedding_type": "absolute",
     "prefix": null,
     "problem_type": null,
@@ -65,6 +67,7 @@
     "sep_token_id": null,
     "task_specific_params": null,
     "temperature": 1.0,
+    "tf_legacy_loss": false,
     "tie_encoder_decoder": false,
     "tie_word_embeddings": true,
     "tokenizer_class": null,
@@ -72,35 +75,46 @@
     "top_p": 1.0,
     "torch_dtype": null,
     "torchscript": false,
-    "transformers_version": "4.15.0",
-    "type_vocab_size": 1,
+    "transformers_version": "4.22.2",
+    "type_vocab_size": 2,
+    "typical_p": 1.0,
     "use_bfloat16": false,
     "use_cache": true,
-    "vocab_size": 50273
+    "vocab_size": 28903
   },
-  "encoder_name": "roberta-base",
+  "encoder_name": "microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract",
   "finetuning_task": [
     "event"
   ],
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
-  "layer": 11,
+  "hier_head_config": null,
+  "label_dictionary": {
+    "event": [
+      "B-AFTER",
+      "B-BEFORE",
+      "B-BEFORE/OVERLAP",
+      "B-OVERLAP",
+      "I-AFTER",
+      "I-BEFORE",
+      "I-BEFORE/OVERLAP",
+      "I-OVERLAP",
+      "O"
+    ]
+  },
+  "layer": 12,
   "model_type": "cnlpt",
-  "num_labels_list": [
-    9
-  ],
   "num_rel_attention_heads": 12,
-  "num_tokens": -1,
   "rel_attention_head_dims": 64,
-  "relations": [
-    false
-  ],
-  "tagger": [
-    true
-  ],
+  "relations": {
+    "event": false
+  },
+  "tagger": {
+    "event": true
+  },
   "tokens": false,
   "torch_dtype": "float32",
-  "transformers_version": "4.15.0",
+  "transformers_version": "4.22.2",
   "use_prior_tasks": false,
-  "vocab_size": 50273
+  "vocab_size": 28903
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:623232360d08a64a4445fb2db5869c2042e5f90ba7496a6bde90e93676d89361
-size 501053685
+oid sha256:4ea51bdfa098b57514ebf75d3e7245564a14f8377c5c7e8ea35521035c92172a
+size 435419517

special_tokens_map.json

@@ -1 +1,17 @@
-{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}, "additional_special_tokens": ["<e>", "</e>", "<a1>", "</a1>", "<a2>", "</a2>", "<cr>", "<neg>"]}
+{
+  "additional_special_tokens": [
+    "<e>",
+    "</e>",
+    "<a1>",
+    "</a1>",
+    "<a2>",
+    "</a2>",
+    "<cr>",
+    "<neg>"
+  ],
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

tokenizer_config.json

@@ -1 +1,26 @@
-{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": true, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "trim_offsets": true, "additional_special_tokens": ["<e>", "</e>", "<a1>", "</a1>", "<a2>", "</a2>", "<cr>", "<neg>"], "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "roberta-base", "tokenizer_class": "RobertaTokenizer"}
+{
+  "add_prefix_space": true,
+  "additional_special_tokens": [
+    "<e>",
+    "</e>",
+    "<a1>",
+    "</a1>",
+    "<a2>",
+    "</a2>",
+    "<cr>",
+    "<neg>"
+  ],
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "name_or_path": "microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract",
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "special_tokens_map_file": null,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}