layoutlmv3-finetuned-Algo_22000Words

This model is a fine-tuned version of microsoft/layoutlmv3-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4487
• Precision: 0.8409
• Recall: 0.8268
• F1: 0.8338
• Accuracy: 0.8727

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: 1e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.03	10	1.8188	0.1983	0.1285	0.1559	0.3227
No log	0.07	20	1.6701	0.1789	0.0950	0.1241	0.3364
No log	0.1	30	1.5675	0.3028	0.1844	0.2292	0.4409
No log	0.14	40	1.4618	0.3803	0.3017	0.3364	0.5182
No log	0.17	50	1.3792	0.4228	0.3520	0.3841	0.5545
No log	0.21	60	1.2919	0.4730	0.3911	0.4281	0.5727
No log	0.24	70	1.2117	0.5513	0.4804	0.5134	0.6364
No log	0.28	80	1.1297	0.6024	0.5587	0.5797	0.6909
No log	0.31	90	1.0708	0.6176	0.5866	0.6017	0.6955
No log	0.35	100	1.0096	0.6095	0.5754	0.5920	0.7091
No log	0.38	110	0.9750	0.5818	0.5363	0.5581	0.6727
No log	0.42	120	0.9478	0.5893	0.5531	0.5706	0.6818
No log	0.45	130	0.8710	0.6494	0.6313	0.6402	0.7318
No log	0.49	140	0.8509	0.6941	0.6592	0.6762	0.7636
No log	0.52	150	0.8065	0.6959	0.6648	0.68	0.7682
No log	0.56	160	0.7702	0.7341	0.7095	0.7216	0.7909
No log	0.59	170	0.7290	0.7529	0.7318	0.7422	0.8045
No log	0.63	180	0.7143	0.7414	0.7207	0.7309	0.7955
No log	0.66	190	0.7161	0.7557	0.7430	0.7493	0.8091
No log	0.7	200	0.6983	0.7443	0.7318	0.7380	0.8
No log	0.73	210	0.6654	0.7771	0.7598	0.7684	0.8273
No log	0.77	220	0.6355	0.7701	0.7486	0.7592	0.8273
No log	0.8	230	0.6380	0.7746	0.7486	0.7614	0.8182
No log	0.84	240	0.6313	0.7784	0.7654	0.7718	0.8273
No log	0.87	250	0.6356	0.7797	0.7709	0.7753	0.8318
No log	0.91	260	0.6125	0.7853	0.7765	0.7809	0.8273
No log	0.94	270	0.6109	0.7943	0.7765	0.7853	0.8364
No log	0.98	280	0.6087	0.7771	0.7598	0.7684	0.8273
No log	1.01	290	0.5677	0.8103	0.7877	0.7989	0.8455
No log	1.05	300	0.5542	0.8057	0.7877	0.7966	0.8409
No log	1.08	310	0.5490	0.8125	0.7989	0.8056	0.85
No log	1.11	320	0.5490	0.8023	0.7933	0.7978	0.8409
No log	1.15	330	0.5632	0.7684	0.7598	0.7640	0.8091
No log	1.18	340	0.5686	0.7809	0.7765	0.7787	0.8182
No log	1.22	350	0.5302	0.7989	0.7989	0.7989	0.8364
No log	1.25	360	0.5101	0.8146	0.8101	0.8123	0.8545
No log	1.29	370	0.5180	0.8023	0.7933	0.7978	0.85
No log	1.32	380	0.5088	0.7966	0.7877	0.7921	0.8409
No log	1.36	390	0.4975	0.8136	0.8045	0.8090	0.8455
No log	1.39	400	0.4914	0.8202	0.8156	0.8179	0.85
No log	1.43	410	0.4870	0.8286	0.8101	0.8192	0.8545
No log	1.46	420	0.4766	0.8171	0.7989	0.8079	0.8455
No log	1.5	430	0.4723	0.8171	0.7989	0.8079	0.8455
No log	1.53	440	0.4675	0.8171	0.7989	0.8079	0.8455
No log	1.57	450	0.4579	0.8305	0.8212	0.8258	0.8636
No log	1.6	460	0.4539	0.8362	0.8268	0.8315	0.8682
No log	1.64	470	0.4505	0.8362	0.8268	0.8315	0.8682
No log	1.67	480	0.4493	0.8514	0.8324	0.8418	0.8773
No log	1.71	490	0.4489	0.8514	0.8324	0.8418	0.8773
0.82	1.74	500	0.4487	0.8409	0.8268	0.8338	0.8727

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1+cu118
• Datasets 2.13.0
• Tokenizers 0.13.3