learning_rate: 2e-05
train_batch_size: 128
eval_batch_size: 128
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 20
training_steps: 6520
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.1507	1.0	652	0.0249	0.8949	0.9105	0.9026	0.9928
0.0256	2.0	1304	0.0189	0.9186	0.9245	0.9215	0.9945
0.0195	3.0	1956	0.0169	0.9237	0.9470	0.9352	0.9952
0.0131	4.0	2608	0.0161	0.9299	0.9499	0.9398	0.9956
0.0114	5.0	3260	0.0149	0.9311	0.9607	0.9457	0.9959
0.01	6.0	3912	0.0146	0.9395	0.9600	0.9497	0.9962
0.0072	7.0	4564	0.0139	0.9480	0.9562	0.9521	0.9965
0.0065	8.0	5216	0.0133	0.9431	0.9655	0.9542	0.9966
0.0059	9.0	5868	0.0134	0.9501	0.9640	0.9570	0.9968
0.0042	10.0	6520	0.0138	0.9505	0.9655	0.9579	0.9969

luohuashijieyoufengjun/ner_based_bert-base-chinese_withBadcase_replaceSpace

ner_based_bert-base-chinese_withBadcase_replaceSpace