metadata
license: cc-by-nc-sa-4.0
base_model: microsoft/layoutlmv3-base
tags:
- generated_from_trainer
datasets:
- doc_lay_net-small
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: Layoutlmv3-finetuned-DocLayNet-test
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: doc_lay_net-small
type: doc_lay_net-small
config: DocLayNet_2022.08_processed_on_2023.01
split: test
args: DocLayNet_2022.08_processed_on_2023.01
metrics:
- name: Precision
type: precision
value: 0.6647646219686163
- name: Recall
type: recall
value: 0.6763425253991292
- name: F1
type: f1
value: 0.6705035971223021
- name: Accuracy
type: accuracy
value: 0.8582839474362278
Layoutlmv3-finetuned-DocLayNet-test
This model is a fine-tuned version of microsoft/layoutlmv3-base on the doc_lay_net-small dataset. It achieves the following results on the evaluation set:
- Loss: 0.8293
- Precision: 0.6648
- Recall: 0.6763
- F1: 0.6705
- Accuracy: 0.8583
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
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 1.5039 | 0.3660 | 250 | 1.1856 | 0.1597 | 0.2785 | 0.2030 | 0.5852 |
| 0.8176 | 0.7321 | 500 | 0.6027 | 0.4143 | 0.5506 | 0.4728 | 0.8651 |
| 0.5533 | 1.0981 | 750 | 0.6755 | 0.5946 | 0.6266 | 0.6102 | 0.8649 |
| 0.4021 | 1.4641 | 1000 | 0.6233 | 0.6017 | 0.6646 | 0.6316 | 0.8804 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1