sataliulan

checkpoint 无法加载

1

我使用BiLSTMCRF做ner任务，embedding为cn-Bert-base-wwm, 模型开始训练时，Trainer设置save_path 为"./save_model" ,训练完成后生成文件：best_BiLSTMCRF_f_2021-07-06-12-14-12-018221; 随后我想加载现有训练结果继续训练，遂编写callback 为CheckPointCallback(save_path='./save_model/best_BiLSTMCRF_f_2021-07-06-12-14-12-018221') 并在trainer中设置callbacks=[callback]; 在训练报错：/fastNLP/core/callback.py line 949 in on_train_begin, model.load_statea-dict(states['model']) TypeError: 'BiLSTMCRF' object is not subscriptable. 不知道是哪里配置出了问题，请排查一下

看到rasa_chinese 引入了PaddleNLP，不知道是否有支持pytorch的打算，毕竟pytorch的生态相对庞大

训练数据问题

1

“ UserWarning: Failed to use example '患感冒咋治呀' to train MITIE entity extractor. Example will be skipped.Error: Invalid entity {'end': 3, 'entity': 'disease', 'start': 1, 'value': '感冒'} in example '患感冒咋治呀': entities...

migrate keras2 code to keras3.0 with pytorch backend, got inplace error

4

hi，there I tried to migrate my keras 2.15 code to keras 3.0 (just like the source code below), the code ran well in keras 2 with backend tensorflow 2.15, but...