layoutlmv3-cordv2

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: 0.3190
• Precision: 0.8830
• Recall: 0.8864
• F1: 0.8847
• Accuracy: 0.9201

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: 1500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.3333	100	1.5910	0.5469	0.4954	0.5199	0.5926
No log	0.6667	200	1.0875	0.6533	0.6685	0.6608	0.7540
No log	1.0	300	0.8007	0.7629	0.7782	0.7705	0.8144
No log	1.3333	400	0.6082	0.8099	0.8130	0.8114	0.8471
1.2178	1.6667	500	0.5302	0.8222	0.8184	0.8203	0.8568
1.2178	2.0	600	0.5046	0.8300	0.8261	0.8280	0.8624
1.2178	2.3333	700	0.4537	0.8498	0.8485	0.8492	0.8785
1.2178	2.6667	800	0.4270	0.8566	0.8586	0.8576	0.8870
1.2178	3.0	900	0.3938	0.8680	0.8686	0.8683	0.8980
0.4128	3.3333	1000	0.3926	0.8765	0.8717	0.8741	0.9031
0.4128	3.6667	1100	0.3403	0.8644	0.8717	0.8680	0.9087
0.4128	4.0	1200	0.3326	0.8861	0.8841	0.8851	0.9159
0.4128	4.3333	1300	0.3223	0.8824	0.8872	0.8848	0.9201
0.4128	4.6667	1400	0.3175	0.8852	0.8879	0.8866	0.9210
0.274	5.0	1500	0.3190	0.8830	0.8864	0.8847	0.9201

Framework versions

• Transformers 4.43.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1