layoutlmv3-final-v1

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: 0.0822
• Precision: 0.9789
• Recall: 0.9839
• F1: 0.9814
• Accuracy: 0.9852

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	2.56	100	1.7056	0.4839	0.5464	0.5133	0.6494
No log	5.13	200	0.4856	0.8255	0.8821	0.8528	0.9217
No log	7.69	300	0.2319	0.9027	0.9355	0.9188	0.9575
No log	10.26	400	0.1437	0.9652	0.9778	0.9715	0.9800
0.9248	12.82	500	0.1204	0.967	0.9748	0.9709	0.9770
0.9248	15.38	600	0.1025	0.9711	0.9808	0.9759	0.9790
0.9248	17.95	700	0.0971	0.9789	0.9829	0.9809	0.9826
0.9248	20.51	800	0.0885	0.9819	0.9849	0.9834	0.9846
0.9248	23.08	900	0.0858	0.9789	0.9839	0.9814	0.9857
0.0697	25.64	1000	0.0822	0.9789	0.9839	0.9814	0.9852

Framework versions

• Transformers 4.30.0.dev0
• Pytorch 1.8.0+cu101
• Datasets 2.12.0
• Tokenizers 0.13.3