metadata

license: cc-by-nc-sa-4.0
base_model: microsoft/layoutlmv3-base
tags:
  - generated_from_trainer
datasets:
  - funsd
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: layoutlmv3-finetuned-funsd
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: funsd
          type: funsd
          config: funsd
          split: test
          args: funsd
        metrics:
          - name: Precision
            type: precision
            value: 0.7467652495378928
          - name: Recall
            type: recall
            value: 0.8027819175360159
          - name: F1
            type: f1
            value: 0.7737610725401005
          - name: Accuracy
            type: accuracy
            value: 0.8188517770117675

layoutlmv3-finetuned-funsd

This model is a fine-tuned version of microsoft/layoutlmv3-base on the funsd dataset. It achieves the following results on the evaluation set:

Loss: 0.5984
Precision: 0.7468
Recall: 0.8028
F1: 0.7738
Accuracy: 0.8189

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

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.67	100	1.0197	0.5025	0.5981	0.5462	0.6622
No log	1.34	200	0.6833	0.6203	0.7238	0.6680	0.7608
No log	2.01	300	0.6237	0.6401	0.7794	0.7030	0.7846
No log	2.68	400	0.6028	0.6892	0.7392	0.7133	0.7771
0.8343	3.36	500	0.5948	0.7175	0.7884	0.7512	0.7991
0.8343	4.03	600	0.5953	0.7135	0.8028	0.7555	0.7961
0.8343	4.7	700	0.5925	0.7354	0.7953	0.7642	0.8174
0.8343	5.37	800	0.6055	0.7397	0.7933	0.7656	0.8134
0.8343	6.04	900	0.5940	0.7535	0.8077	0.7797	0.8199
0.3468	6.71	1000	0.5984	0.7468	0.8028	0.7738	0.8189

Framework versions

Transformers 4.35.0.dev0
Pytorch 2.1.0+cu121
Datasets 2.14.6
Tokenizers 0.14.1