EElayoutlmv3_jordyvl_rvl_cdip_100_examples_per_class_2023-09-29_lte_test

This model is a fine-tuned version of microsoft/layoutlmv3-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.5042
• Accuracy: 0.75

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 2
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 24
• total_train_batch_size: 48
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 60

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.96	16	2.6827	0.1375
No log	1.98	33	2.5024	0.24
No log	3.0	50	2.3456	0.315
No log	3.96	66	2.1650	0.3625
No log	4.98	83	1.9151	0.475
No log	6.0	100	1.6499	0.6025
No log	6.96	116	1.4450	0.6425
No log	7.98	133	1.2701	0.6825
No log	9.0	150	1.1825	0.6725
No log	9.96	166	1.0343	0.7275
No log	10.98	183	1.0277	0.72
No log	12.0	200	0.9242	0.7625
No log	12.96	216	0.9228	0.745
No log	13.98	233	1.0066	0.7125
No log	15.0	250	0.9636	0.75
No log	15.96	266	0.9258	0.7475
No log	16.98	283	1.0153	0.745
No log	18.0	300	1.0909	0.7375
No log	18.96	316	1.1108	0.735
No log	19.98	333	1.0873	0.74
No log	21.0	350	1.0968	0.75
No log	21.96	366	1.1242	0.7625
No log	22.98	383	1.1795	0.7575
No log	24.0	400	1.1529	0.7625
No log	24.96	416	1.1900	0.76
No log	25.98	433	1.2364	0.75
No log	27.0	450	1.2554	0.7675
No log	27.96	466	1.2810	0.75
No log	28.98	483	1.3241	0.755
5.5136	30.0	500	1.3343	0.7625
5.5136	30.96	516	1.3430	0.7575
5.5136	31.98	533	1.3808	0.7525
5.5136	33.0	550	1.3886	0.7575
5.5136	33.96	566	1.3628	0.7625
5.5136	34.98	583	1.3966	0.745
5.5136	36.0	600	1.3708	0.7625
5.5136	36.96	616	1.4044	0.755
5.5136	37.98	633	1.4421	0.755
5.5136	39.0	650	1.4101	0.7575
5.5136	39.96	666	1.4206	0.755
5.5136	40.98	683	1.4098	0.7725
5.5136	42.0	700	1.4874	0.745
5.5136	42.96	716	1.5017	0.75
5.5136	43.98	733	1.4326	0.77
5.5136	45.0	750	1.4896	0.7575
5.5136	45.96	766	1.4124	0.7725
5.5136	46.98	783	1.4505	0.765
5.5136	48.0	800	1.4823	0.755
5.5136	48.96	816	1.4516	0.765
5.5136	49.98	833	1.4879	0.7575
5.5136	51.0	850	1.4876	0.755
5.5136	51.96	866	1.4850	0.755
5.5136	52.98	883	1.5151	0.7575
5.5136	54.0	900	1.5031	0.76
5.5136	54.96	916	1.4955	0.7575
5.5136	55.98	933	1.5084	0.75
5.5136	57.0	950	1.5053	0.75
5.5136	57.6	960	1.5042	0.75

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu117
• Datasets 2.13.1
• Tokenizers 0.13.3