Modelpublic

dwitidibyajyoti/fine_tune_layoutmlv3_model

sourceHugging Facecc-by-nc-sa-4.0updated 3y agoView on Hugging Face

0likes12downloads

Model Card

test

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: 0.2763
—Precision: 0.5109
—Recall: 0.6026
—F1: 0.5529
—Accuracy: 0.9222

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
—trainbatchsize: 2
—evalbatchsize: 2
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	8.33	100	0.6800	0.3371	0.3846	0.3593	0.7682
No log	16.67	200	0.3088	0.5204	0.6538	0.5795	0.9156
No log	25.0	300	0.2142	0.5326	0.6282	0.5765	0.9305
No log	33.33	400	0.2301	0.5795	0.6538	0.6145	0.9288
0.4115	41.67	500	0.2426	0.5618	0.6410	0.5988	0.9272
0.4115	50.0	600	0.4171	0.6190	0.6667	0.6420	0.8924
0.4115	58.33	700	0.2265	0.5393	0.6154	0.5749	0.9371
0.4115	66.67	800	0.2869	0.5506	0.6282	0.5868	0.9156
0.4115	75.0	900	0.2633	0.5568	0.6282	0.5904	0.9272
0.0231	83.33	1000	0.2763	0.5109	0.6026	0.5529	0.9222

Framework versions

—Transformers 4.33.0.dev0
—Pytorch 2.0.1+cu118
—Datasets 2.14.4
—Tokenizers 0.13.3