|
--- |
|
language: |
|
- en |
|
license: mit |
|
base_model: microsoft/deberta-v3-base |
|
tags: |
|
- nycu-112-2-datamining-hw2 |
|
- generated_from_trainer |
|
datasets: |
|
- DandinPower/review_onlytitleandtext |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: deberta-v3-base-otat |
|
results: |
|
- task: |
|
name: Text Classification |
|
type: text-classification |
|
dataset: |
|
name: DandinPower/review_onlytitleandtext |
|
type: DandinPower/review_onlytitleandtext |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.639 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# deberta-v3-base-otat |
|
|
|
This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co./microsoft/deberta-v3-base) on the DandinPower/review_onlytitleandtext dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.4437 |
|
- Accuracy: 0.639 |
|
- Macro F1: 0.6399 |
|
|
|
## 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: 4.5e-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 |
|
- lr_scheduler_warmup_steps: 1500 |
|
- num_epochs: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Macro F1 | |
|
|:-------------:|:-----:|:-----:|:---------------:|:--------:|:--------:| |
|
| 0.9984 | 0.14 | 500 | 0.9957 | 0.5819 | 0.5794 | |
|
| 1.0009 | 0.29 | 1000 | 0.9064 | 0.6161 | 0.6222 | |
|
| 0.9462 | 0.43 | 1500 | 0.9272 | 0.6047 | 0.5906 | |
|
| 0.9037 | 0.57 | 2000 | 0.9866 | 0.5817 | 0.5750 | |
|
| 0.8923 | 0.71 | 2500 | 0.8666 | 0.6124 | 0.5898 | |
|
| 0.905 | 0.86 | 3000 | 0.8855 | 0.5996 | 0.5745 | |
|
| 0.9017 | 1.0 | 3500 | 0.8521 | 0.6276 | 0.6258 | |
|
| 0.8487 | 1.14 | 4000 | 0.8540 | 0.6309 | 0.6292 | |
|
| 0.8042 | 1.29 | 4500 | 0.8534 | 0.6323 | 0.6294 | |
|
| 0.8165 | 1.43 | 5000 | 0.8350 | 0.6347 | 0.6389 | |
|
| 0.8224 | 1.57 | 5500 | 0.8687 | 0.6321 | 0.6279 | |
|
| 0.7799 | 1.71 | 6000 | 0.8810 | 0.6316 | 0.6298 | |
|
| 0.7354 | 1.86 | 6500 | 0.8719 | 0.639 | 0.6346 | |
|
| 0.8026 | 2.0 | 7000 | 0.8829 | 0.6159 | 0.6154 | |
|
| 0.6818 | 2.14 | 7500 | 0.9274 | 0.6383 | 0.6408 | |
|
| 0.6704 | 2.29 | 8000 | 0.9327 | 0.6401 | 0.6377 | |
|
| 0.6498 | 2.43 | 8500 | 0.8786 | 0.6367 | 0.6414 | |
|
| 0.6956 | 2.57 | 9000 | 0.9165 | 0.6374 | 0.6320 | |
|
| 0.6729 | 2.71 | 9500 | 0.9929 | 0.6116 | 0.6153 | |
|
| 0.6963 | 2.86 | 10000 | 0.8843 | 0.6397 | 0.6418 | |
|
| 0.6795 | 3.0 | 10500 | 0.9204 | 0.6471 | 0.6492 | |
|
| 0.536 | 3.14 | 11000 | 1.0496 | 0.641 | 0.6447 | |
|
| 0.5212 | 3.29 | 11500 | 1.0836 | 0.6466 | 0.6466 | |
|
| 0.5278 | 3.43 | 12000 | 1.0635 | 0.6377 | 0.6420 | |
|
| 0.5631 | 3.57 | 12500 | 1.0144 | 0.6436 | 0.6449 | |
|
| 0.4899 | 3.71 | 13000 | 1.1613 | 0.6416 | 0.6420 | |
|
| 0.509 | 3.86 | 13500 | 1.0841 | 0.6446 | 0.6442 | |
|
| 0.5176 | 4.0 | 14000 | 1.0819 | 0.639 | 0.6426 | |
|
| 0.3587 | 4.14 | 14500 | 1.3046 | 0.6401 | 0.6412 | |
|
| 0.4342 | 4.29 | 15000 | 1.3250 | 0.6371 | 0.6394 | |
|
| 0.3358 | 4.43 | 15500 | 1.4140 | 0.6387 | 0.6395 | |
|
| 0.3773 | 4.57 | 16000 | 1.4286 | 0.6399 | 0.6416 | |
|
| 0.4173 | 4.71 | 16500 | 1.4825 | 0.6393 | 0.6396 | |
|
| 0.4072 | 4.86 | 17000 | 1.4357 | 0.6393 | 0.6405 | |
|
| 0.3743 | 5.0 | 17500 | 1.4437 | 0.639 | 0.6399 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.39.3 |
|
- Pytorch 2.2.2+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
