init

config.json

@@ -0,0 +1,54 @@
+{
+  "_name_or_path": "xlm-roberta-large",
+  "architectures": [
+    "XLMRobertaForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "O",
+    "1": "B-location",
+    "2": "I-location",
+    "3": "B-group",
+    "4": "B-corporation",
+    "5": "B-person",
+    "6": "B-work of art",
+    "7": "B-product",
+    "8": "I-person",
+    "9": "I-work of art",
+    "10": "I-corporation",
+    "11": "I-group",
+    "12": "I-product"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "B-corporation": 4,
+    "B-group": 3,
+    "B-location": 1,
+    "B-person": 5,
+    "B-product": 7,
+    "B-work of art": 6,
+    "I-corporation": 10,
+    "I-group": 11,
+    "I-location": 2,
+    "I-person": 8,
+    "I-product": 12,
+    "I-work of art": 9,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "output_past": true,
+  "pad_token_id": 1,
+  "type_vocab_size": 1,
+  "vocab_size": 250002
+}

parameter.json

@@ -0,0 +1 @@
+{"dataset": ["wnut2017"], "transformers_model": "xlm-roberta-large", "random_seed": 1234, "lr": 1e-05, "total_step": 5000, "warmup_step": 700, "weight_decay": 1e-07, "batch_size": 32, "max_seq_length": 128, "fp16": false, "max_grad_norm": 1.0, "lower_case": false}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3cdcd0c727fbcf0a9bf3d93b33ed5b4c2ff1420fe55619e7ba0616cf3a677dc
+size 2235581546

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}

test_bc5cdr_span.json

@@ -0,0 +1 @@
+{"valid": {"f1": 0.0, "recall": 0.0, "precision": 0.0, "summary": ""}, "test": {"f1": 0.0, "recall": 0.0, "precision": 0.0, "summary": ""}}

test_bionlp2004_span.json

@@ -0,0 +1 @@
+{"valid": {"f1": 0.0, "recall": 0.0, "precision": 0.0, "summary": ""}}

test_conll2003_span.json

@@ -0,0 +1 @@
+{"valid": {"f1": 89.02770233568712, "recall": 89.51392681594757, "precision": 88.54673149648839, "accuracy": 98.24006868024662, "summary": "              precision    recall  f1-score   support\n\n      entity       0.89      0.90      0.89      3662\n\n   micro avg       0.89      0.90      0.89      3662\n   macro avg       0.89      0.90      0.89      3662\nweighted avg       0.89      0.90      0.89      3662\n"}, "test": {"f1": 85.69667077681873, "recall": 85.01529051987767, "precision": 86.38906152889993, "accuracy": 97.76114574121605, "summary": "              precision    recall  f1-score   support\n\n      entity       0.86      0.85      0.86      3270\n\n   micro avg       0.86      0.85      0.86      3270\n   macro avg       0.86      0.85      0.86      3270\nweighted avg       0.86      0.85      0.86      3270\n"}}

test_fin_span.json

@@ -0,0 +1 @@
+{"valid": {"f1": 20.074349442379187, "recall": 13.366336633663368, "precision": 40.298507462686565, "accuracy": 97.76627360751607, "summary": "              precision    recall  f1-score   support\n\n      entity       0.40      0.13      0.20       202\n\n   micro avg       0.40      0.13      0.20       202\n   macro avg       0.40      0.13      0.20       202\nweighted avg       0.40      0.13      0.20       202\n"}}

test_ontonotes5_span.json

@@ -0,0 +1 @@
+{"valid": {"f1": 41.13397548161121, "recall": 57.30405611466911, "precision": 32.08127027488475, "accuracy": 95.39802007051804, "summary": "              precision    recall  f1-score   support\n\n      entity       0.32      0.57      0.41      3279\n\n   micro avg       0.32      0.57      0.41      3279\n   macro avg       0.32      0.57      0.41      3279\nweighted avg       0.32      0.57      0.41      3279\n"}, "test": {"f1": 41.891274554591135, "recall": 56.43076923076923, "precision": 33.30911732655285, "accuracy": 95.6830572447054, "summary": "              precision    recall  f1-score   support\n\n      entity       0.33      0.56      0.42      3250\n\n   micro avg       0.33      0.56      0.42      3250\n   macro avg       0.33      0.56      0.42      3250\nweighted avg       0.33      0.56      0.42      3250\n"}}

test_panx_dataset-en_span.json

@@ -0,0 +1 @@
+{"valid": {"f1": 55.62348926489407, "recall": 62.174189446916714, "precision": 50.32158476974531, "accuracy": 82.891302726753, "summary": "              precision    recall  f1-score   support\n\n      entity       0.50      0.62      0.56      9438\n\n   micro avg       0.50      0.62      0.56      9438\n   macro avg       0.50      0.62      0.56      9438\nweighted avg       0.50      0.62      0.56      9438\n"}, "test": {"f1": 54.52257658616851, "recall": 61.56284153005465, "precision": 48.92729957439416, "accuracy": 82.83985276686006, "summary": "              precision    recall  f1-score   support\n\n      entity       0.49      0.62      0.55      9150\n\n   micro avg       0.49      0.62      0.55      9150\n   macro avg       0.49      0.62      0.55      9150\nweighted avg       0.49      0.62      0.55      9150\n"}}

test_wnut2017.json

@@ -0,0 +1 @@
+{"valid": {"f1": 66.36363636363636, "recall": 61.1244019138756, "precision": 72.58522727272727, "accuracy": 96.4483482910063, "summary": "              precision    recall  f1-score   support\n\n corporation       0.38      0.38      0.38        34\n       group       0.41      0.41      0.41        39\n    location       0.69      0.62      0.65        74\n      person       0.86      0.77      0.81       470\n     product       0.51      0.41      0.46       114\n work of art       0.57      0.28      0.37       105\n\n   micro avg       0.73      0.61      0.66       836\n   macro avg       0.57      0.48      0.51       836\nweighted avg       0.72      0.61      0.66       836\n"}, "test": {"f1": 58.575197889182064, "recall": 51.532033426183844, "precision": 67.84841075794621, "accuracy": 95.61209918213683, "summary": "              precision    recall  f1-score   support\n\n corporation       0.37      0.44      0.40        66\n       group       0.60      0.33      0.43       165\n    location       0.73      0.64      0.68       150\n      person       0.78      0.69      0.73       428\n     product       0.46      0.23      0.31       127\n work of art       0.69      0.35      0.46       141\n\n   micro avg       0.68      0.52      0.59      1077\n   macro avg       0.60      0.45      0.50      1077\nweighted avg       0.67      0.52      0.57      1077\n"}}

test_wnut2017_span.json

@@ -0,0 +1 @@
+{"valid": {"f1": 75.09778357235984, "recall": 68.89952153110048, "precision": 82.52148997134671, "accuracy": 97.14212971803195, "summary": "              precision    recall  f1-score   support\n\n      entity       0.83      0.69      0.75       836\n\n   micro avg       0.83      0.69      0.75       836\n   macro avg       0.83      0.69      0.75       836\nweighted avg       0.83      0.69      0.75       836\n"}, "test": {"f1": 68.32363828662083, "recall": 59.98142989786444, "precision": 79.36117936117935, "accuracy": 96.25686961789779, "summary": "              precision    recall  f1-score   support\n\n      entity       0.79      0.60      0.68      1077\n\n   micro avg       0.79      0.60      0.68      1077\n   macro avg       0.79      0.60      0.68      1077\nweighted avg       0.79      0.60      0.68      1077\n"}}

tokenizer_config.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "model_max_length": 512, "name_or_path": "xlm-roberta-large"}