|
--- |
|
base_model: microsoft/mdeberta-v3-base |
|
library_name: transformers |
|
license: mit |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: scenario-non-kd-scr-ner-full-mdeberta_data-univner_full44 |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# scenario-non-kd-scr-ner-full-mdeberta_data-univner_full44 |
|
|
|
This model is a fine-tuned version of [microsoft/mdeberta-v3-base](https://huggingface.co/microsoft/mdeberta-v3-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3003 |
|
- Precision: 0.6230 |
|
- Recall: 0.5993 |
|
- F1: 0.6109 |
|
- Accuracy: 0.9631 |
|
|
|
## 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: 3e-05 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 32 |
|
- seed: 44 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 30 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-------:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 0.3129 | 0.2910 | 500 | 0.2430 | 0.3687 | 0.2001 | 0.2594 | 0.9351 | |
|
| 0.201 | 0.5821 | 1000 | 0.1893 | 0.3603 | 0.3711 | 0.3656 | 0.9430 | |
|
| 0.1493 | 0.8731 | 1500 | 0.1664 | 0.4946 | 0.4279 | 0.4588 | 0.9519 | |
|
| 0.1081 | 1.1641 | 2000 | 0.1566 | 0.5297 | 0.5299 | 0.5298 | 0.9563 | |
|
| 0.0881 | 1.4552 | 2500 | 0.1487 | 0.5472 | 0.5748 | 0.5607 | 0.9581 | |
|
| 0.0825 | 1.7462 | 3000 | 0.1487 | 0.5918 | 0.5183 | 0.5526 | 0.9594 | |
|
| 0.0721 | 2.0373 | 3500 | 0.1490 | 0.5893 | 0.5660 | 0.5774 | 0.9613 | |
|
| 0.0454 | 2.3283 | 4000 | 0.1648 | 0.5981 | 0.5498 | 0.5730 | 0.9609 | |
|
| 0.0457 | 2.6193 | 4500 | 0.1633 | 0.5882 | 0.5934 | 0.5908 | 0.9617 | |
|
| 0.046 | 2.9104 | 5000 | 0.1505 | 0.6074 | 0.5959 | 0.6016 | 0.9627 | |
|
| 0.0302 | 3.2014 | 5500 | 0.1771 | 0.6159 | 0.5788 | 0.5968 | 0.9620 | |
|
| 0.0249 | 3.4924 | 6000 | 0.1871 | 0.6064 | 0.5751 | 0.5903 | 0.9615 | |
|
| 0.0271 | 3.7835 | 6500 | 0.1806 | 0.6146 | 0.5882 | 0.6011 | 0.9628 | |
|
| 0.0235 | 4.0745 | 7000 | 0.1966 | 0.6161 | 0.5804 | 0.5977 | 0.9626 | |
|
| 0.0152 | 4.3655 | 7500 | 0.2110 | 0.6071 | 0.5887 | 0.5978 | 0.9621 | |
|
| 0.0165 | 4.6566 | 8000 | 0.1978 | 0.6008 | 0.6174 | 0.6090 | 0.9620 | |
|
| 0.0164 | 4.9476 | 8500 | 0.2096 | 0.6029 | 0.5750 | 0.5886 | 0.9611 | |
|
| 0.011 | 5.2386 | 9000 | 0.2174 | 0.6055 | 0.6027 | 0.6041 | 0.9626 | |
|
| 0.0101 | 5.5297 | 9500 | 0.2234 | 0.5919 | 0.6080 | 0.5999 | 0.9615 | |
|
| 0.0109 | 5.8207 | 10000 | 0.2246 | 0.6148 | 0.5975 | 0.6060 | 0.9623 | |
|
| 0.0099 | 6.1118 | 10500 | 0.2228 | 0.6115 | 0.6164 | 0.6139 | 0.9626 | |
|
| 0.0062 | 6.4028 | 11000 | 0.2401 | 0.6099 | 0.6060 | 0.6079 | 0.9623 | |
|
| 0.0073 | 6.6938 | 11500 | 0.2560 | 0.6161 | 0.5897 | 0.6026 | 0.9621 | |
|
| 0.0082 | 6.9849 | 12000 | 0.2488 | 0.6008 | 0.5914 | 0.5960 | 0.9614 | |
|
| 0.0049 | 7.2759 | 12500 | 0.2573 | 0.6155 | 0.5832 | 0.5989 | 0.9620 | |
|
| 0.0057 | 7.5669 | 13000 | 0.2583 | 0.6320 | 0.5882 | 0.6093 | 0.9628 | |
|
| 0.0058 | 7.8580 | 13500 | 0.2601 | 0.6040 | 0.6188 | 0.6113 | 0.9623 | |
|
| 0.0044 | 8.1490 | 14000 | 0.2676 | 0.5962 | 0.6006 | 0.5984 | 0.9616 | |
|
| 0.0039 | 8.4400 | 14500 | 0.2747 | 0.6194 | 0.5930 | 0.6059 | 0.9624 | |
|
| 0.004 | 8.7311 | 15000 | 0.2796 | 0.6080 | 0.5776 | 0.5924 | 0.9614 | |
|
| 0.0044 | 9.0221 | 15500 | 0.2836 | 0.6095 | 0.5875 | 0.5983 | 0.9623 | |
|
| 0.0028 | 9.3132 | 16000 | 0.2907 | 0.6315 | 0.5891 | 0.6095 | 0.9631 | |
|
| 0.003 | 9.6042 | 16500 | 0.2962 | 0.6212 | 0.5787 | 0.5992 | 0.9626 | |
|
| 0.0038 | 9.8952 | 17000 | 0.2864 | 0.6232 | 0.5823 | 0.6021 | 0.9625 | |
|
| 0.0029 | 10.1863 | 17500 | 0.2912 | 0.6240 | 0.5892 | 0.6061 | 0.9623 | |
|
| 0.0023 | 10.4773 | 18000 | 0.2990 | 0.6344 | 0.5728 | 0.6020 | 0.9625 | |
|
| 0.0028 | 10.7683 | 18500 | 0.2953 | 0.6186 | 0.5965 | 0.6073 | 0.9628 | |
|
| 0.0021 | 11.0594 | 19000 | 0.2989 | 0.6216 | 0.5988 | 0.6100 | 0.9630 | |
|
| 0.0017 | 11.3504 | 19500 | 0.3025 | 0.6161 | 0.6057 | 0.6108 | 0.9631 | |
|
| 0.0023 | 11.6414 | 20000 | 0.2973 | 0.6148 | 0.6057 | 0.6102 | 0.9629 | |
|
| 0.0021 | 11.9325 | 20500 | 0.3003 | 0.6230 | 0.5993 | 0.6109 | 0.9631 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.1.1+cu121 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.19.1 |
|
|
