learning_rate: 5e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 8
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: linear
num_epochs: 20
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Accuracy	Precision	Recall	F1	Validation Loss
No log	1.0	13	0.5463	0.5179	0.5463	0.4339	1.0286
No log	2.0	26	0.4100	0.1876	0.4100	0.2461	1.0251
No log	3.0	39	0.5086	0.6698	0.5086	0.4356	0.8710
No log	4.0	52	0.4802	0.6966	0.4802	0.3729	1.3227
No log	5.0	65	0.4691	0.7161	0.4691	0.3548	1.0735
No log	6.0	78	0.5927	0.6874	0.5927	0.5628	0.8102
No log	7.0	91	0.5402	0.7032	0.5402	0.4823	1.3396
No log	8.0	104	0.6661	0.6753	0.6661	0.6340	0.7542
No log	9.0	117	0.6234	0.6935	0.6234	0.6047	0.8814
No log	10.0	130	0.6633	0.6732	0.6633	0.6574	0.7494
No log	11.0	143	0.6567	0.6597	0.6567	0.6520	0.7748
No log	12.0	156	0.6606	0.6596	0.6606	0.6552	0.7600
No log	13.0	169	0.6624	0.6744	0.6624	0.6567	0.7976
No log	14.0	182	0.6667	0.6668	0.6667	0.6619	0.7685
No log	15.0	195	0.6452	0.6778	0.6452	0.6361	0.8573
No log	16.0	208	0.6536	0.6721	0.6536	0.6466	0.8498
No log	17.0	221	0.6545	0.6625	0.6545	0.6501	0.8457
No log	18.0	234	0.6570	0.6602	0.6570	0.6523	0.8187
No log	18.48	240	0.6579	0.6594	0.6579	0.6534	0.8215

yemen2016/memo3_indirect_speech

memo3_indirect_speech