20NG_ALBERT_5E
This model is a fine-tuned version of albert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2209
- Accuracy: 0.6067
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.8602 | 0.07 | 50 | 2.5794 | 0.2133 |
| 2.3635 | 0.14 | 100 | 2.0956 | 0.38 |
| 2.1526 | 0.21 | 150 | 1.9011 | 0.4467 |
| 1.9014 | 0.28 | 200 | 1.6340 | 0.5067 |
| 1.6736 | 0.35 | 250 | 1.5457 | 0.5467 |
| 1.5563 | 0.42 | 300 | 1.5041 | 0.5533 |
| 1.4338 | 0.49 | 350 | 1.3933 | 0.5933 |
| 1.3348 | 0.56 | 400 | 1.4123 | 0.54 |
| 1.2879 | 0.64 | 450 | 1.3352 | 0.6333 |
| 1.2864 | 0.71 | 500 | 1.3027 | 0.62 |
| 1.2162 | 0.78 | 550 | 1.2734 | 0.6267 |
| 1.1786 | 0.85 | 600 | 1.2695 | 0.5933 |
| 1.1702 | 0.92 | 650 | 1.2379 | 0.5933 |
| 1.2338 | 0.99 | 700 | 1.2209 | 0.6067 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.1
- Datasets 2.8.0
- Tokenizers 0.13.2