Initial Commit

README.md

@@ -0,0 +1,86 @@
+---
+license: mit
+base_model: microsoft/mdeberta-v3-base
+tags:
+- generated_from_trainer
+metrics:
+- precision
+- recall
+- f1
+- accuracy
+model-index:
+- name: scenario-TCR-NER_data-univner_half
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# scenario-TCR-NER_data-univner_half
+
+This model is a fine-tuned version of [microsoft/mdeberta-v3-base](https://huggingface.co/microsoft/mdeberta-v3-base) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1170
+- Precision: 0.8494
+- Recall: 0.8655
+- F1: 0.8574
+- Accuracy: 0.9842
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 3e-05
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 30
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 0.1168        | 0.58  | 500   | 0.0625          | 0.8182    | 0.8512 | 0.8344 | 0.9825   |
+| 0.0433        | 1.17  | 1000  | 0.0594          | 0.8396    | 0.8632 | 0.8512 | 0.9843   |
+| 0.0305        | 1.75  | 1500  | 0.0677          | 0.8296    | 0.8703 | 0.8495 | 0.9836   |
+| 0.0213        | 2.33  | 2000  | 0.0761          | 0.8253    | 0.8833 | 0.8533 | 0.9839   |
+| 0.0185        | 2.91  | 2500  | 0.0738          | 0.8600    | 0.8612 | 0.8606 | 0.9850   |
+| 0.012         | 3.5   | 3000  | 0.0784          | 0.8374    | 0.8572 | 0.8471 | 0.9835   |
+| 0.0124        | 4.08  | 3500  | 0.0832          | 0.8363    | 0.8704 | 0.8530 | 0.9843   |
+| 0.0095        | 4.66  | 4000  | 0.0806          | 0.8423    | 0.8713 | 0.8565 | 0.9845   |
+| 0.008         | 5.24  | 4500  | 0.1049          | 0.8218    | 0.8625 | 0.8417 | 0.9823   |
+| 0.0071        | 5.83  | 5000  | 0.0879          | 0.8420    | 0.8632 | 0.8525 | 0.9842   |
+| 0.0068        | 6.41  | 5500  | 0.0918          | 0.8507    | 0.8733 | 0.8619 | 0.9846   |
+| 0.0058        | 6.99  | 6000  | 0.0951          | 0.8488    | 0.8667 | 0.8577 | 0.9845   |
+| 0.0047        | 7.58  | 6500  | 0.0991          | 0.8467    | 0.8651 | 0.8558 | 0.9842   |
+| 0.0047        | 8.16  | 7000  | 0.1025          | 0.8603    | 0.8573 | 0.8588 | 0.9845   |
+| 0.0043        | 8.74  | 7500  | 0.1020          | 0.8473    | 0.8678 | 0.8574 | 0.9845   |
+| 0.0031        | 9.32  | 8000  | 0.1085          | 0.8437    | 0.8582 | 0.8509 | 0.9842   |
+| 0.0038        | 9.91  | 8500  | 0.1082          | 0.8602    | 0.8440 | 0.8520 | 0.9839   |
+| 0.0024        | 10.49 | 9000  | 0.1163          | 0.8533    | 0.8544 | 0.8539 | 0.9838   |
+| 0.0038        | 11.07 | 9500  | 0.1139          | 0.8528    | 0.8567 | 0.8548 | 0.9843   |
+| 0.0024        | 11.66 | 10000 | 0.1130          | 0.8619    | 0.8476 | 0.8547 | 0.9841   |
+| 0.0024        | 12.24 | 10500 | 0.1170          | 0.8494    | 0.8655 | 0.8574 | 0.9842   |
+
+
+### Framework versions
+
+- Transformers 4.33.3
+- Pytorch 2.1.1+cu121
+- Datasets 2.14.5
+- Tokenizers 0.13.3

config.json

@@ -0,0 +1,53 @@
+{
+  "_name_or_path": "microsoft/mdeberta-v3-base",
+  "architectures": [
+    "DebertaV2ForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4",
+    "5": "LABEL_5",
+    "6": "LABEL_6"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4,
+    "LABEL_5": 5,
+    "LABEL_6": 6
+  },
+  "layer_norm_eps": 1e-07,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta-v2",
+  "norm_rel_ebd": "layer_norm",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "position_buckets": 256,
+  "relative_attention": true,
+  "share_att_key": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.33.3",
+  "type_vocab_size": 0,
+  "vocab_size": 251000
+}

eval_result_ner.json

@@ -0,0 +1 @@
+{"zh_gsd": {"precision": 0.8458646616541353, "recall": 0.8800521512385919, "f1": 0.8626198083067093, "accuracy": 0.9806859806859807}, "zh_gsdsimp": {"precision": 0.8640406607369758, "recall": 0.891218872870249, "f1": 0.8774193548387097, "accuracy": 0.9823509823509824}, "hr_set": {"precision": 0.9101508916323731, "recall": 0.9458303635067712, "f1": 0.9276476756378887, "accuracy": 0.990560593569662}, "da_ddt": {"precision": 0.8681818181818182, "recall": 0.854586129753915, "f1": 0.8613303269447576, "accuracy": 0.9895240945824604}, "en_ewt": {"precision": 0.836555360281195, "recall": 0.875, "f1": 0.8553459119496856, "accuracy": 0.9850579750567797}, "pt_bosque": {"precision": 0.8723404255319149, "recall": 0.8436213991769548, "f1": 0.8577405857740587, "accuracy": 0.9858716128097377}, "sr_set": {"precision": 0.9494117647058824, "recall": 0.9527744982290437, "f1": 0.9510901591043017, "accuracy": 0.990631293231766}, "sk_snk": {"precision": 0.8242229367631297, "recall": 0.8404371584699454, "f1": 0.8322510822510822, "accuracy": 0.9757380653266332}, "sv_talbanken": {"precision": 0.7763713080168776, "recall": 0.9387755102040817, "f1": 0.8498845265588915, "accuracy": 0.9969082789419443}}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b17284dbdc47a46c624b01a42698c89dde812636572a5924ce8bda92d62475f5
+size 1112965930

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2af879047a05ce26c181c49037265b598df6baec29f1ae74753378b1c252b119
+size 4536