README.md

metadata

tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: layoutlmv3-er-ner
    results: []

layoutlmv3-er-ner

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

• Loss: 0.1905
• Precision: 0.6189
• Recall: 0.6591
• F1: 0.6384
• Accuracy: 0.9529

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: 5e-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
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	22	0.2818	0.3420	0.2801	0.3079	0.9064
No log	2.0	44	0.1898	0.5280	0.4540	0.4882	0.9380
No log	3.0	66	0.1729	0.5992	0.6153	0.6071	0.9479
No log	4.0	88	0.1880	0.6179	0.6040	0.6109	0.9457
No log	5.0	110	0.1845	0.5607	0.5941	0.5769	0.9451
No log	6.0	132	0.1869	0.6071	0.6294	0.6181	0.9501
No log	7.0	154	0.1838	0.6288	0.6492	0.6388	0.9517
No log	8.0	176	0.1928	0.5973	0.6167	0.6068	0.9503
No log	9.0	198	0.1899	0.6179	0.6634	0.6398	0.9521
No log	10.0	220	0.1905	0.6189	0.6591	0.6384	0.9529

Framework versions

• Transformers 4.20.0.dev0
• Pytorch 1.11.0
• Datasets 2.1.0
• Tokenizers 0.12.1