metadata
base_model: textattack/albert-base-v2-imdb
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: albert-tune
results: []
albert-tune
This model is a fine-tuned version of textattack/albert-base-v2-imdb on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8791
- Accuracy: 0.6857
- F1: 0.7039
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
- lr_scheduler_warmup_steps: 500
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 1.8887 | 1.0 | 53 | 1.8444 | 0.1929 | 0.1619 |
| 1.6055 | 2.0 | 106 | 1.5226 | 0.5929 | 0.5821 |
| 1.2048 | 3.0 | 159 | 1.1546 | 0.5857 | 0.5779 |
| 0.7243 | 4.0 | 212 | 0.9967 | 0.6214 | 0.6173 |
| 0.6455 | 5.0 | 265 | 0.9122 | 0.6857 | 0.6941 |
| 0.9166 | 6.0 | 318 | 0.8791 | 0.6857 | 0.7039 |
| 0.505 | 7.0 | 371 | 1.0556 | 0.6643 | 0.6665 |
Framework versions
- Transformers 4.43.3
- Pytorch 2.4.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1