|
--- |
|
license: mit |
|
base_model: microsoft/deberta-v3-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- precision |
|
- recall |
|
- f1 |
|
model-index: |
|
- name: DeBERT_50K_steps |
|
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. --> |
|
|
|
# DeBERT_50K_steps |
|
|
|
This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co/microsoft/deberta-v3-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0169 |
|
- Accuracy: 0.9941 |
|
- Precision: 0.7649 |
|
- Recall: 0.5670 |
|
- F1: 0.6512 |
|
- Hamming: 0.0059 |
|
|
|
## 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: 2e-05 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 16 |
|
- 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: 50000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Hamming | |
|
|:-------------:|:-----:|:-----:|:---------------:|:--------:|:---------:|:------:|:------:|:-------:| |
|
| 0.2014 | 0.02 | 2500 | 0.0451 | 0.9902 | 0.0 | 0.0 | 0.0 | 0.0098 | |
|
| 0.0373 | 0.04 | 5000 | 0.0297 | 0.9913 | 0.6879 | 0.2003 | 0.3102 | 0.0087 | |
|
| 0.0286 | 0.06 | 7500 | 0.0250 | 0.9921 | 0.6965 | 0.3329 | 0.4505 | 0.0079 | |
|
| 0.0253 | 0.08 | 10000 | 0.0233 | 0.9925 | 0.7038 | 0.4010 | 0.5109 | 0.0075 | |
|
| 0.0234 | 0.1 | 12500 | 0.0217 | 0.9928 | 0.7085 | 0.4382 | 0.5415 | 0.0072 | |
|
| 0.0223 | 0.12 | 15000 | 0.0208 | 0.9930 | 0.7229 | 0.4559 | 0.5591 | 0.0070 | |
|
| 0.0213 | 0.14 | 17500 | 0.0205 | 0.9931 | 0.7255 | 0.4696 | 0.5701 | 0.0069 | |
|
| 0.0206 | 0.16 | 20000 | 0.0196 | 0.9933 | 0.7325 | 0.4990 | 0.5936 | 0.0067 | |
|
| 0.0203 | 0.18 | 22500 | 0.0191 | 0.9935 | 0.7368 | 0.5125 | 0.6045 | 0.0065 | |
|
| 0.0196 | 0.2 | 25000 | 0.0188 | 0.9935 | 0.7354 | 0.5209 | 0.6098 | 0.0065 | |
|
| 0.0195 | 0.22 | 27500 | 0.0185 | 0.9936 | 0.7415 | 0.5335 | 0.6205 | 0.0064 | |
|
| 0.019 | 0.24 | 30000 | 0.0183 | 0.9936 | 0.7437 | 0.5296 | 0.6186 | 0.0064 | |
|
| 0.0189 | 0.26 | 32500 | 0.0180 | 0.9938 | 0.7585 | 0.5304 | 0.6243 | 0.0062 | |
|
| 0.0187 | 0.28 | 35000 | 0.0178 | 0.9938 | 0.7630 | 0.5342 | 0.6284 | 0.0062 | |
|
| 0.0184 | 0.3 | 37500 | 0.0175 | 0.9939 | 0.7626 | 0.5457 | 0.6362 | 0.0061 | |
|
| 0.0182 | 0.32 | 40000 | 0.0174 | 0.9939 | 0.7621 | 0.5451 | 0.6356 | 0.0061 | |
|
| 0.0179 | 0.34 | 42500 | 0.0172 | 0.9940 | 0.7594 | 0.5563 | 0.6422 | 0.0060 | |
|
| 0.0178 | 0.36 | 45000 | 0.0171 | 0.9940 | 0.7553 | 0.5633 | 0.6453 | 0.0060 | |
|
| 0.0177 | 0.38 | 47500 | 0.0170 | 0.9941 | 0.7623 | 0.5680 | 0.6510 | 0.0059 | |
|
| 0.0175 | 0.4 | 50000 | 0.0169 | 0.9941 | 0.7649 | 0.5670 | 0.6512 | 0.0059 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.0.dev0 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.7.1 |
|
- Tokenizers 0.14.1 |
|
|
