learning_rate: 1e-05
train_batch_size: 16
eval_batch_size: 16
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: linear
lr_scheduler_warmup_ratio: 0.08
training_steps: 2000
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.3496	1.0	110	0.1492	0.0	0.0	0.0	0.0
0.1157	2.0	220	0.0945	0.2584	0.0292	0.0524	0.0486
0.0665	3.0	330	0.0610	0.4959	0.3029	0.3761	0.4101
0.0469	4.0	440	0.0429	0.7739	0.5120	0.6163	0.5743
0.0326	5.0	550	0.0352	0.8982	0.7047	0.7898	0.7512
0.0274	6.0	660	0.0306	0.9099	0.7681	0.8330	0.7998
0.0232	7.0	770	0.0284	0.9331	0.7782	0.8487	0.8115
0.0197	8.0	880	0.0277	0.9288	0.8099	0.8653	0.8397
0.0191	9.0	990	0.0288	0.9227	0.7719	0.8406	0.8095
0.0158	10.0	1100	0.0280	0.9324	0.8035	0.8632	0.8338
0.0144	11.0	1210	0.0285	0.9275	0.8112	0.8654	0.8416
0.0143	12.0	1320	0.0289	0.9316	0.8112	0.8672	0.8445
0.0136	13.0	1430	0.0297	0.9390	0.8188	0.8747	0.8484
0.0136	14.0	1540	0.0284	0.9339	0.8061	0.8653	0.8367
0.0125	15.0	1650	0.0285	0.9344	0.7947	0.8589	0.8280
0.012	16.0	1760	0.0293	0.9374	0.8162	0.8726	0.8455
0.0113	17.0	1870	0.0292	0.9305	0.7807	0.8491	0.8154

Giannis17/layoutlmv3-finetuned-invoice_ConControl_v304

layoutlmv3-finetuned-invoice_ConControl_v304