训练完后找不到模型

![输入图片说明](https://foruda.gitee.com/images/1689558346315904075/a8a42d37_13211094.png "屏幕截图")

训练完后 找不到模型，不知道哪里的问题，  而且这个数据训练特别快，也没报错，不知道哪里的问题。  日志如下：

C:\Users\zjhe\PycharmProjects\study\bert_bilstm_crf_ner_pytorch-master\torch_ner\output\20230717085727
2023-07-17 08:57:27,447 - __main__ - INFO - available device: cpu，count_gpu: 0
2023-07-17 08:57:27,447 - __main__ - INFO - ====================== Start Data Pre-processing ======================
2023-07-17 08:57:27,447 - root - INFO - loading labels info from train file and dump in C:\Users\zjhe\PycharmProjects\study\bert_bilstm_crf_ner_pytorch-master\torch_ner\output\20230717085727
2023-07-17 08:57:27,559 - __main__ - INFO - loading labels successful! the size is 15, label is: I-LNAME,I-LABEL,B-DATE,B-LOC,B-ORG,B-LABEL,I-LOC,I-DATE,I-CW,B-LNAME,I-ORG,I-FNAME,B-CW,O,B-FNAME
2023-07-17 08:57:27,560 - __main__ - INFO - loading label2id and id2label dictionary successful!
Some weights of the model checkpoint at C:\Users\zjhe\PycharmProjects\study\bert_bilstm_crf_ner_pytorch-master\torch_ner\bert-base-chinese were not used when initializing BERT_BiLSTM_CRF: ['cls.seq_relationship.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.predictions.decoder.weight', 'cls.predictions.transform.dense.bias', 'cls.seq_relationship.weight', 'cls.predictions.bias', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.transform.dense.weight']
- This IS expected if you are initializing BERT_BiLSTM_CRF from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
- This IS NOT expected if you are initializing BERT_BiLSTM_CRF from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
Some weights of BERT_BiLSTM_CRF were not initialized from the model checkpoint at C:\Users\zjhe\PycharmProjects\study\bert_bilstm_crf_ner_pytorch-master\torch_ner\bert-base-chinese and are newly initialized: ['hidden2tag.bias', 'birnn.weight_hh_l0_reverse', 'crf.end_transitions', 'crf.start_transitions', 'hidden2tag.weight', 'birnn.bias_hh_l0_reverse', 'birnn.bias_hh_l0', 'birnn.bias_ih_l0_reverse', 'crf.transitions', 'birnn.weight_ih_l0_reverse', 'birnn.bias_ih_l0', 'birnn.weight_ih_l0', 'birnn.weight_hh_l0']
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
2023-07-17 08:57:28,493 - __main__ - INFO - loading tokenizer、bert_config and bert_bilstm_crf model successful!
2023-07-17 08:57:28,493 - __main__ - INFO - starting load train data and data_loader...
file_path: C:\Users\zjhe\PycharmProjects\study\bert_bilstm_crf_ner_pytorch-master\torch_ner\data\train.txt
get_input_examples,len(lines): 957 ['O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O B-CW I-CW O', '在 重 大 的 人 生 事 件 或 经 历 军 队 体 制 编 制 改 革 时 ， 存 在 焦 虑 性 尤 为 明 显 ， 特 别 是 面 临 家 庭 带 来 的 种 种 困 扰 ， 它 很 容 易 影 响 年 轻 军 官 。']
convert examples:   0%|          | 0/957 [00:00<?, ?it/s]2023-07-17 08:57:28,634 - processor - INFO - ====================================================================== Example ======================================================================
2023-07-17 08:57:28,634 - processor - INFO - guid: 0
2023-07-17 08:57:28,634 - processor - INFO - tokens: ['[CLS]', '早', '在', '5', '月', '1', '5', '日', '，', '武', '汉', '警', '方', '在', '腾', '讯', '手', '机', '管', '家', '、', '腾', '讯', '电', '脑', '管', '家', '及', '腾', '讯', '守', '[SEP]']
2023-07-17 08:57:28,634 - processor - INFO - input_ids: 101 3193 1762 126 3299 122 126 3189 8024 3636 3727 6356 3175 1762 5596 6380 2797 3322 5052 2157 510 5596 6380 4510 5554 5052 2157 1350 5596 6380 2127 102
2023-07-17 08:57:28,634 - processor - INFO - token_type_ids: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
2023-07-17 08:57:28,634 - processor - INFO - attention_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
2023-07-17 08:57:28,634 - processor - INFO - label_ids: 13 13 13 2 7 7 7 7 13 4 10 10 10 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 4 10 10 13
2023-07-17 08:57:28,641 - processor - INFO - ====================================================================== Example ======================================================================
2023-07-17 08:57:28,641 - processor - INFO - guid: 1
2023-07-17 08:57:28,642 - processor - INFO - tokens: ['[CLS]', '1', '1', '月', '7', '日', '，', '我', '们', '踏', '上', '了', '这', '片', '红', '土', '地', '寻', '访', '英', '雄', '[UNK]', '[UNK]', '百', '岁', '老', '兵', '陈', '训', '杨', '：', '[SEP]']
2023-07-17 08:57:28,642 - processor - INFO - input_ids: 101 122 122 3299 128 3189 8024 2769 812 6672 677 749 6821 4275 5273 1759 1765 2192 6393 5739 7413 100 100 4636 2259 5439 1070 7357 6378 3342 8038 102
2023-07-17 08:57:28,642 - processor - INFO - token_type_ids: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
2023-07-17 08:57:28,642 - processor - INFO - attention_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
2023-07-17 08:57:28,642 - processor - INFO - label_ids: 13 2 7 7 7 7 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 5 1 1 1 14 9 0 13 13
2023-07-17 08:57:28,647 - processor - INFO - ====================================================================== Example ======================================================================
2023-07-17 08:57:28,647 - processor - INFO - guid: 2
2023-07-17 08:57:28,647 - processor - INFO - tokens: ['[CLS]', '在', '重', '大', '的', '人', '生', '事', '件', '或', '经', '历', '军', '队', '体', '制', '编', '制', '改', '革', '时', '，', '存', '在', '焦', '虑', '性', '尤', '为', '明', '显', '[SEP]']
2023-07-17 08:57:28,647 - processor - INFO - input_ids: 101 1762 7028 1920 4638 782 4495 752 816 2772 5307 1325 1092 7339 860 1169 5356 1169 3121 7484 3198 8024 2100 1762 4193 5991 2595 2215 711 3209 3227 102
2023-07-17 08:57:28,647 - processor - INFO - token_type_ids: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
2023-07-17 08:57:28,647 - processor - INFO - attention_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
2023-07-17 08:57:28,647 - processor - INFO - label_ids: 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13
convert examples: 100%|██████████| 957/957 [00:05<00:00, 183.91it/s]
2023-07-17 08:57:33,833 - __main__ - INFO - loading train data_set and data_loader successful!
2023-07-17 08:57:33,833 - __main__ - INFO - ====================== End Data Pre-processing ======================
C:\ProgramData\Anaconda3\envs\zjhe\lib\site-packages\transformers\optimization.py:309: FutureWarning: This implementation of AdamW is deprecated and will be removed in a future version. Use the PyTorch implementation torch.optim.AdamW instead, or set `no_deprecation_warning=True` to disable this warning
  FutureWarning,
2023-07-17 08:57:33,838 - __main__ - INFO - loading AdamW optimizer、Warmup LinearSchedule and calculate optimizer parameter successful!
2023-07-17 08:57:33,838 - __main__ - INFO - ====================== Running training ======================
2023-07-17 08:57:33,838 - __main__ - INFO - Num Examples:  3, Num Batch Step: 1, Num Epochs: 2, Num scheduler steps：2
Epoch:   0%|          | 0/2 [00:00<?, ?it/s]2023-07-17 08:57:33,839 - __main__ - INFO - ########[Epoch: 0/2]########

DataLoader:   0%|          | 0/1 [00:00<?, ?it/s]2023-07-17 08:57:33,840 - __main__ - INFO - ####[Step: 0/1]####

DataLoader: 100%|██████████| 1/1 [00:01<00:00,  1.12s/it]
Epoch:  50%|█████     | 1/2 [00:01<00:01,  1.12s/it]2023-07-17 08:57:34,962 - __main__ - INFO - ########[Epoch: 1/2]########

DataLoader:   0%|          | 0/1 [00:00<?, ?it/s]2023-07-17 08:57:34,964 - __main__ - INFO - ####[Step: 0/1]####

DataLoader: 100%|██████████| 1/1 [00:01<00:00,  1.31s/it]
Epoch: 100%|██████████| 2/2 [00:02<00:00,  1.21s/it]
2023-07-17 08:57:36,271 - __main__ - INFO - NER model training successful!!!

Process finished with exit code 0

chenzhou/bert_bilstm_crf_ner_pytorch

Content Risk Flag

Comments (9)

chenzhou/bert_bilstm_crf_ner_pytorch .gitee-modal { width: 500px !important; }

Content Risk Flag