在cpu上进行训练模型时，entity的precision，recall和F1 都是0

[zhouxiaofang]

校友您好，能帮忙看下源码适配CPU运行吗？

2023-05-26 01:39:08 - INFO: dict_items([('dataset', 'resume-zh'), ('dist_emb_size', 20), ('type_emb_size', 20), ('lstm_hid_size', 512), ('conv_hid_size', 96), ('bert_hid_size', 768), ('biaffine_size', 512), ('ffnn_hid_size', 288), ('dilation', [1, 2, 3]), ('emb_dropout', 0.5), ('conv_dropout', 0.5), ('out_dropout', 0.33), ('epochs', 10), ('batch_size', 16), ('learning_rate', 0.001), ('weight_decay', 0), ('clip_grad_norm', 5.0), ('bert_name', 'bert-base-chinese'), ('bert_learning_rate', 5e-06), ('warm_factor', 0.1), ('use_bert_last_4_layers', False), ('seed', 123), ('config', 'config/resume-zh.json'), ('device', 'cpu'), ('fp16', False), ('use_precision_alignment', False)]) 2023-05-26 01:39:08 - INFO: Loading Data 2023-05-26 01:39:08 - INFO: +-----------+-----------+----------+ | resume-zh | sentences | entities | +-----------+-----------+----------+ | train | 3819 | 13438 | | dev | 463 | 1497 | | test | 477 | 1630 | +-----------+-----------+----------+ 2023-05-26 01:39:26 - INFO: Building Model Some weights of the model checkpoint at bert-base-chinese were not used when initializing BertModel: ['cls.predictions.transform.LayerNorm.bias', 'cls.predictions.transform.dense.bias', 'cls.predictions.decoder.weight', 'cls.seq_relationship.bias', 'cls.predictions.bias', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.transform.dense.weight']

• This IS expected if you are initializing BertModel 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 BertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). 2023-05-26 01:39:28 - INFO: Epoch: 0 ---start training--- s222 x type: torch.float32 s222 x type: torch.Size([15, 552, 180, 180]) /data/zhejiang/yusl/anaconda3/envs/w2ner_teco/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1308: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use zero_division parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result)) 2023-05-26 01:44:07 - INFO: EVAL Label F1 [0.993111 0. 0. 0. 0. 0. 0. 0.
  0. 0. ] 2023-05-26 01:44:07 - INFO: +--------+--------+-----------+--------+ | EVAL 0 | F1 | Precision | Recall | +--------+--------+-----------+--------+ | Label | 0.0993 | 0.0987 | 0.0999 | | Entity | 0.0000 | 0.0000 | 0.0000 |

[fanjiaxin23]

@zhouxiaofang 你能修好它吗？我有同样的问题，知道如何解决吗？