Model description

More information needed

Intended uses & limitations

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 32
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 256
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_with_restarts
lr_scheduler_warmup_ratio: 0.1
num_epochs: 50
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	3	1.5852	0.1932
No log	2.0	6	1.5784	0.3182
No log	3.0	9	1.5374	0.4318
1.3768	4.0	12	1.4629	0.4091
1.3768	5.0	15	1.2222	0.5341
1.3768	6.0	18	1.2437	0.5455
0.942	7.0	21	1.2428	0.5341
0.942	8.0	24	1.1751	0.5341
0.942	9.0	27	1.1279	0.5795
0.6265	10.0	30	0.9868	0.6477
0.6265	11.0	33	0.9661	0.6364
0.6265	12.0	36	0.9892	0.6136
0.6265	13.0	39	0.8716	0.6818
0.5106	14.0	42	0.8010	0.7273
0.5106	15.0	45	0.8596	0.6818
0.5106	16.0	48	0.8257	0.6932
0.4183	17.0	51	0.8190	0.7045
0.4183	18.0	54	0.7628	0.7273
0.4183	19.0	57	0.7802	0.7159
0.3267	20.0	60	0.7753	0.75
0.3267	21.0	63	0.7771	0.7386
0.3267	22.0	66	0.7770	0.75
0.3267	23.0	69	0.7941	0.7273
0.3008	24.0	72	0.7921	0.7273
0.3008	25.0	75	0.7899	0.7386
0.3008	26.0	78	0.7849	0.75
0.2795	27.0	81	0.7891	0.75
0.2795	28.0	84	0.7973	0.7386
0.2795	29.0	87	0.8068	0.7386
0.2526	30.0	90	0.8088	0.7386
0.2526	31.0	93	0.8098	0.7386
0.2526	32.0	96	0.8096	0.7386
0.2526	33.0	99	0.8095	0.7386
0.2544	33.3810	100	0.8094	0.7386