Closed Dhaizei closed 9 months ago
反正我是把metric_value =0.9,进行赋值了,最后并没有印象到模型的训练,就是不知道这个metrics在哪里计算的宏观f1,反正是没有,也没有找到,气死。
python run_train.py --device gpu --logging_steps 10 --save_steps 100 --eval_steps 100 --seed 1000 --model_name_or_path utc-xbase --output_dir ./checkpoint/model_best_12 --dataset_path ./data/data_12 --max_seq_length 512 --per_device_train_batch_size 2 --per_device_eval_batch_size 2 --gradient_accumulation_steps 8 --num_train_epochs 20 --learning_rate 1e-5 --do_train --do_eval --do_export --export_model_dir ./checkpoint/model_best_12 --overwrite_output_dir --disable_tqdm True --metric_for_best_model macro_f1 --load_best_model_at_end True --save_total_limit 1 --save_plm --fp16 True
反正我是把metric_value =0.9,进行赋值了,最后并没有印象到模型的训练,就是不知道这个metrics在哪里计算的宏观f1,反正是没有,也没有找到,气死。
试一下这个兄弟:
python run_train.py --device gpu --logging_steps 10 --save_steps 100 --eval_steps 100 --seed 1000 --model_name_or_path utc-xbase --output_dir ./checkpoint/model_best_12 --dataset_path ./data/data_12 --max_seq_length 512 --per_device_train_batch_size 2 --per_device_eval_batch_size 2 --gradient_accumulation_steps 8 --num_train_epochs 20 --learning_rate 1e-5 --do_train --do_eval --do_export --export_model_dir ./checkpoint/model_best_12 --overwrite_output_dir --disable_tqdm True --metric_for_best_model macro_f1 --load_best_model_at_end True --save_total_limit 1 --save_plm --fp16 True
请提出你的问题
执行脚本 python run_train.py \ --device gpu \ --logging_steps 10 \ --save_steps 100 \ --eval_steps 100 \ --seed 1000 \ --model_name_or_path utc-base \ --output_dir ./checkpoint/model_best \ --dataset_path ./data/ \ --max_seq_length 512 \ --per_device_train_batch_size 2 \ --per_device_eval_batch_size 2 \ --gradient_accumulation_steps 8 \ --num_train_epochs 20 \ --learning_rate 1e-5 \ --do_train \ --do_eval \ --do_export \ --export_model_dir ./checkpoint/model_best \ --overwrite_output_dir \ --disable_tqdm True \ --metric_for_best_model macro_f1 \ --load_best_model_at_end True \ --save_total_limit 1 \ --save_plm
metrics: {'eval_runtime': 0.019, 'eval_samples_per_second': 0.0, 'eval_steps_per_second': 0.0, 'epoch': 0.6144} metric_to_check: eval_macro_f1 Traceback (most recent call last): File "D:\work\test\PaddleNLP\applications\zero_shot_text_classification\run_train.py", line 154, in
main()
File "D:\work\test\PaddleNLP\applications\zero_shot_text_classification\run_train.py", line 133, in main
train_results = trainer.train(resume_from_checkpoint=training_args.resume_from_checkpoint)
File "D:\software\anaconda3\envs\nlp\lib\site-packages\paddlenlp\trainer\trainer.py", line 888, in train
self._maybe_log_save_evaluate(tr_loss, model, epoch, ignore_keys_for_eval, inputs=inputs)
File "D:\software\anaconda3\envs\nlp\lib\site-packages\paddlenlp\trainer\trainer.py", line 1065, in _maybe_log_save_evaluate
self._save_checkpoint(model, metrics=metrics)
File "D:\software\anaconda3\envs\nlp\lib\site-packages\paddlenlp\trainer\trainer.py", line 1842, in _save_checkpoint
metric_value = metrics[metric_to_check]
KeyError: 'eval_macro_f1'