|
|
--- |
|
|
license: mit |
|
|
base_model: microsoft/deberta-v3-base |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- accuracy |
|
|
model-index: |
|
|
- name: Intent-classification-DeBERTa-model-Ashuv2 |
|
|
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. --> |
|
|
|
|
|
# Intent-classification-DeBERTa-model-Ashuv2 |
|
|
|
|
|
This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co/microsoft/deberta-v3-base) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.1666 |
|
|
- Accuracy: 0.9012 |
|
|
|
|
|
## 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: 5e-05 |
|
|
- train_batch_size: 8 |
|
|
- eval_batch_size: 8 |
|
|
- seed: 42 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 5 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
|
| 1.7819 | 0.12 | 10 | 1.7969 | 0.3665 | |
|
|
| 1.7316 | 0.25 | 20 | 1.6351 | 0.3975 | |
|
|
| 1.4279 | 0.37 | 30 | 1.2845 | 0.5776 | |
|
|
| 1.0181 | 0.49 | 40 | 0.8974 | 0.7143 | |
|
|
| 0.7285 | 0.62 | 50 | 0.6361 | 0.7640 | |
|
|
| 0.7265 | 0.74 | 60 | 0.4886 | 0.8509 | |
|
|
| 0.5108 | 0.86 | 70 | 0.3599 | 0.9006 | |
|
|
| 0.4413 | 0.99 | 80 | 0.2510 | 0.8944 | |
|
|
| 0.3556 | 1.11 | 90 | 0.2156 | 0.9130 | |
|
|
| 0.2884 | 1.23 | 100 | 0.2777 | 0.8944 | |
|
|
| 0.1914 | 1.36 | 110 | 0.2518 | 0.8944 | |
|
|
| 0.5051 | 1.48 | 120 | 0.2118 | 0.9130 | |
|
|
| 0.1151 | 1.6 | 130 | 0.1957 | 0.9130 | |
|
|
| 0.1745 | 1.73 | 140 | 0.2052 | 0.8820 | |
|
|
| 0.1987 | 1.85 | 150 | 0.2053 | 0.8882 | |
|
|
| 0.2467 | 1.98 | 160 | 0.1945 | 0.8944 | |
|
|
| 0.3075 | 2.1 | 170 | 0.2680 | 0.8944 | |
|
|
| 0.1732 | 2.22 | 180 | 0.2642 | 0.8882 | |
|
|
| 0.1627 | 2.35 | 190 | 0.1915 | 0.9068 | |
|
|
| 0.1766 | 2.47 | 200 | 0.1708 | 0.9130 | |
|
|
| 0.2563 | 2.59 | 210 | 0.1691 | 0.8944 | |
|
|
| 0.189 | 2.72 | 220 | 0.1941 | 0.9130 | |
|
|
| 0.1696 | 2.84 | 230 | 0.1907 | 0.9130 | |
|
|
| 0.1865 | 2.96 | 240 | 0.4247 | 0.9130 | |
|
|
| 0.3183 | 3.09 | 250 | 0.2251 | 0.8944 | |
|
|
| 0.185 | 3.21 | 260 | 0.2289 | 0.8882 | |
|
|
| 0.1636 | 3.33 | 270 | 0.1887 | 0.8944 | |
|
|
| 0.2432 | 3.46 | 280 | 0.2055 | 0.8882 | |
|
|
| 0.1518 | 3.58 | 290 | 0.2703 | 0.8944 | |
|
|
| 0.2371 | 3.7 | 300 | 0.2638 | 0.8944 | |
|
|
| 0.1742 | 3.83 | 310 | 0.2309 | 0.8944 | |
|
|
| 0.2269 | 3.95 | 320 | 0.2208 | 0.8882 | |
|
|
| 0.1404 | 4.07 | 330 | 0.2156 | 0.8820 | |
|
|
| 0.1056 | 4.2 | 340 | 0.2192 | 0.9006 | |
|
|
| 0.164 | 4.32 | 350 | 0.2282 | 0.9068 | |
|
|
| 0.1419 | 4.44 | 360 | 0.2380 | 0.9068 | |
|
|
| 0.1164 | 4.57 | 370 | 0.2438 | 0.9006 | |
|
|
| 0.2167 | 4.69 | 380 | 0.2429 | 0.9006 | |
|
|
| 0.1244 | 4.81 | 390 | 0.2363 | 0.8820 | |
|
|
| 0.2341 | 4.94 | 400 | 0.2344 | 0.8882 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.38.2 |
|
|
- Pytorch 2.1.2+cpu |
|
|
- Datasets 2.1.0 |
|
|
- Tokenizers 0.15.2 |
|
|
|
