<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
perioli_vgm_v4.1
This model is a fine-tuned version of microsoft/layoutlmv3-base on the sroie dataset. It achieves the following results on the evaluation set:
- Loss: 0.1084
- Precision: 0.6981
- Recall: 0.6789
- F1: 0.6884
- Accuracy: 0.9881
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 | 0.98 | 100 | 0.0735 | 0.7849 | 0.6697 | 0.7228 | 0.9901 |
No log | 1.96 | 200 | 0.0688 | 0.6638 | 0.7064 | 0.6844 | 0.9905 |
No log | 2.94 | 300 | 0.0827 | 0.6636 | 0.6697 | 0.6667 | 0.9877 |
No log | 3.92 | 400 | 0.0717 | 0.7170 | 0.6972 | 0.7070 | 0.9901 |
0.0088 | 4.9 | 500 | 0.0715 | 0.6667 | 0.6422 | 0.6542 | 0.9881 |
0.0088 | 5.88 | 600 | 0.0868 | 0.7059 | 0.6606 | 0.6825 | 0.9877 |
0.0088 | 6.86 | 700 | 0.0947 | 0.6491 | 0.6789 | 0.6637 | 0.9881 |
0.0088 | 7.84 | 800 | 0.0796 | 0.7103 | 0.6972 | 0.7037 | 0.9894 |
0.0088 | 8.82 | 900 | 0.0777 | 0.7196 | 0.7064 | 0.7130 | 0.9905 |
0.0027 | 9.8 | 1000 | 0.0958 | 0.7075 | 0.6881 | 0.6977 | 0.9877 |
0.0027 | 10.78 | 1100 | 0.1044 | 0.71 | 0.6514 | 0.6794 | 0.9874 |
0.0027 | 11.76 | 1200 | 0.1140 | 0.6981 | 0.6789 | 0.6884 | 0.9874 |
0.0027 | 12.75 | 1300 | 0.1106 | 0.6981 | 0.6789 | 0.6884 | 0.9881 |
0.0027 | 13.73 | 1400 | 0.1069 | 0.6981 | 0.6789 | 0.6884 | 0.9881 |
0.0009 | 14.71 | 1500 | 0.1084 | 0.6981 | 0.6789 | 0.6884 | 0.9881 |
Framework versions
- Transformers 4.28.0.dev0
- Pytorch 2.0.0+cu118
- Datasets 2.2.2
- Tokenizers 0.13.3