<!-- 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. -->
rhenus_v2.0
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: 0.2385
- Precision: 0.8050
- Recall: 0.8086
- F1: 0.8068
- Accuracy: 0.9704
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: 6000
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 1.25 | 250 | 0.1729 | 0.7857 | 0.7973 | 0.7915 | 0.9687 |
0.0306 | 2.5 | 500 | 0.2302 | 0.7665 | 0.7882 | 0.7772 | 0.9614 |
0.0306 | 3.75 | 750 | 0.1761 | 0.7438 | 0.8086 | 0.7748 | 0.9648 |
0.0216 | 5.0 | 1000 | 0.1935 | 0.7566 | 0.8063 | 0.7807 | 0.9663 |
0.0216 | 6.25 | 1250 | 0.2136 | 0.7684 | 0.7928 | 0.7804 | 0.9663 |
0.016 | 7.5 | 1500 | 0.2358 | 0.7702 | 0.7894 | 0.7796 | 0.9653 |
0.016 | 8.75 | 1750 | 0.2213 | 0.7627 | 0.8007 | 0.7812 | 0.9656 |
0.0101 | 10.0 | 2000 | 0.2142 | 0.7840 | 0.8222 | 0.8027 | 0.9690 |
0.0101 | 11.25 | 2250 | 0.2440 | 0.7710 | 0.7701 | 0.7705 | 0.9637 |
0.0077 | 12.5 | 2500 | 0.2447 | 0.7984 | 0.7848 | 0.7915 | 0.9680 |
0.0077 | 13.75 | 2750 | 0.2034 | 0.8045 | 0.8109 | 0.8077 | 0.9696 |
0.0057 | 15.0 | 3000 | 0.2531 | 0.8086 | 0.7848 | 0.7966 | 0.9693 |
0.0057 | 16.25 | 3250 | 0.2410 | 0.7789 | 0.7939 | 0.7863 | 0.9671 |
0.0054 | 17.5 | 3500 | 0.2406 | 0.7930 | 0.7939 | 0.7934 | 0.9677 |
0.0054 | 18.75 | 3750 | 0.2512 | 0.7815 | 0.7860 | 0.7837 | 0.9656 |
0.0034 | 20.0 | 4000 | 0.2149 | 0.8047 | 0.8120 | 0.8083 | 0.9690 |
0.0034 | 21.25 | 4250 | 0.2195 | 0.8149 | 0.8075 | 0.8111 | 0.9701 |
0.0022 | 22.5 | 4500 | 0.2185 | 0.7954 | 0.8143 | 0.8047 | 0.9690 |
0.0022 | 23.75 | 4750 | 0.2305 | 0.8086 | 0.8086 | 0.8086 | 0.9701 |
0.0018 | 25.0 | 5000 | 0.2266 | 0.7921 | 0.8199 | 0.8058 | 0.9693 |
0.0018 | 26.25 | 5250 | 0.2275 | 0.7952 | 0.8177 | 0.8063 | 0.9706 |
0.0013 | 27.5 | 5500 | 0.2322 | 0.8049 | 0.8177 | 0.8112 | 0.9716 |
0.0013 | 28.75 | 5750 | 0.2340 | 0.8054 | 0.8154 | 0.8104 | 0.9714 |
0.0011 | 30.0 | 6000 | 0.2385 | 0.8050 | 0.8086 | 0.8068 | 0.9704 |
Framework versions
- Transformers 4.28.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.2.2
- Tokenizers 0.13.2