|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- stereoset |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: bert-base-uncased_stereoset_finetuned |
|
results: |
|
- task: |
|
name: Text Classification |
|
type: text-classification |
|
dataset: |
|
name: stereoset |
|
type: stereoset |
|
config: intersentence |
|
split: validation |
|
args: intersentence |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.7260596546310832 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# bert-base-uncased_stereoset_finetuned |
|
|
|
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the stereoset dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.3464 |
|
- Accuracy: 0.7261 |
|
|
|
## 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: 128 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 10 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| No log | 0.21 | 5 | 0.6832 | 0.5565 | |
|
| No log | 0.42 | 10 | 0.6945 | 0.4741 | |
|
| No log | 0.62 | 15 | 0.6659 | 0.6224 | |
|
| No log | 0.83 | 20 | 0.6337 | 0.6758 | |
|
| No log | 1.04 | 25 | 0.6019 | 0.6695 | |
|
| No log | 1.25 | 30 | 0.5797 | 0.7096 | |
|
| No log | 1.46 | 35 | 0.5562 | 0.7166 | |
|
| No log | 1.67 | 40 | 0.5497 | 0.7363 | |
|
| No log | 1.88 | 45 | 0.5382 | 0.7418 | |
|
| No log | 2.08 | 50 | 0.5356 | 0.7418 | |
|
| No log | 2.29 | 55 | 0.5690 | 0.7316 | |
|
| No log | 2.5 | 60 | 0.5778 | 0.7418 | |
|
| No log | 2.71 | 65 | 0.5695 | 0.7386 | |
|
| No log | 2.92 | 70 | 0.5765 | 0.7386 | |
|
| No log | 3.12 | 75 | 0.6079 | 0.7363 | |
|
| No log | 3.33 | 80 | 0.6919 | 0.7370 | |
|
| No log | 3.54 | 85 | 0.7396 | 0.7339 | |
|
| No log | 3.75 | 90 | 0.7109 | 0.7339 | |
|
| No log | 3.96 | 95 | 0.7246 | 0.7308 | |
|
| No log | 4.17 | 100 | 0.7502 | 0.7292 | |
|
| No log | 4.38 | 105 | 0.8222 | 0.7331 | |
|
| No log | 4.58 | 110 | 0.8508 | 0.7268 | |
|
| No log | 4.79 | 115 | 0.8995 | 0.7378 | |
|
| No log | 5.0 | 120 | 0.8797 | 0.7323 | |
|
| No log | 5.21 | 125 | 0.9254 | 0.7370 | |
|
| No log | 5.42 | 130 | 0.9863 | 0.7292 | |
|
| No log | 5.62 | 135 | 1.0044 | 0.7198 | |
|
| No log | 5.83 | 140 | 1.0236 | 0.7339 | |
|
| No log | 6.04 | 145 | 1.0705 | 0.7355 | |
|
| No log | 6.25 | 150 | 1.0734 | 0.7323 | |
|
| No log | 6.46 | 155 | 1.1066 | 0.7300 | |
|
| No log | 6.67 | 160 | 1.1166 | 0.7292 | |
|
| No log | 6.88 | 165 | 1.1258 | 0.7370 | |
|
| No log | 7.08 | 170 | 1.1972 | 0.7300 | |
|
| No log | 7.29 | 175 | 1.1682 | 0.7268 | |
|
| No log | 7.5 | 180 | 1.2221 | 0.7166 | |
|
| No log | 7.71 | 185 | 1.2813 | 0.7151 | |
|
| No log | 7.92 | 190 | 1.3180 | 0.7214 | |
|
| No log | 8.12 | 195 | 1.3224 | 0.7166 | |
|
| No log | 8.33 | 200 | 1.3252 | 0.7135 | |
|
| No log | 8.54 | 205 | 1.3205 | 0.7221 | |
|
| No log | 8.75 | 210 | 1.3266 | 0.7245 | |
|
| No log | 8.96 | 215 | 1.3318 | 0.7206 | |
|
| No log | 9.17 | 220 | 1.3345 | 0.7237 | |
|
| No log | 9.38 | 225 | 1.3378 | 0.7245 | |
|
| No log | 9.58 | 230 | 1.3422 | 0.7261 | |
|
| No log | 9.79 | 235 | 1.3453 | 0.7261 | |
|
| No log | 10.0 | 240 | 1.3464 | 0.7261 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.13.1 |
|
- Datasets 2.9.0 |
|
- Tokenizers 0.13.2 |
|
|
