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_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 2000
num_epochs: 20

Training Loss	Epoch	Step	Validation Loss
3.8885	1.0	500	2.8574
2.6752	2.0	1000	2.4912
2.4468	3.0	1500	2.3058
2.2929	4.0	2000	2.0936
2.1177	5.0	2500	1.8581
1.9363	6.0	3000	1.7443
1.7838	7.0	3500	1.6501
1.7087	8.0	4000	1.6039
1.5766	9.0	4500	1.4060
1.5136	10.0	5000	1.3654
1.4449	11.0	5500	1.3400
1.4177	12.0	6000	1.3540
1.3819	13.0	6500	1.2696
1.3320	14.0	7000	1.2256
1.2864	15.0	7500	1.1872
1.2523	16.0	8000	1.1975
1.2380	17.0	8500	1.1712
1.2173	18.0	9000	1.1291
1.1903	19.0	9500	1.1228
1.1677	20.0	10000	1.1135

kiadamson/detr-fashion-detection

detr-fashion-detection