metadata
base_model: Fsoft-AIC/videberta-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: videberta-base_1024
results: []
videberta-base_1024
This model is a fine-tuned version of Fsoft-AIC/videberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5634
- Accuracy: 0.75
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: 0.0003
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.18
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6979 | 0.1 | 50 | 0.6724 | 0.75 |
| 0.6979 | 0.21 | 100 | 0.6272 | 0.75 |
| 0.6979 | 0.31 | 150 | 0.5727 | 0.75 |
| 0.6979 | 0.41 | 200 | 0.5659 | 0.75 |
| 0.6979 | 0.52 | 250 | 0.5675 | 0.75 |
| 0.6979 | 0.62 | 300 | 0.5633 | 0.75 |
| 0.6979 | 0.72 | 350 | 0.5931 | 0.75 |
| 0.6979 | 0.83 | 400 | 0.5644 | 0.75 |
| 0.6979 | 0.93 | 450 | 0.5633 | 0.75 |
| 0.6979 | 1.03 | 500 | 0.5926 | 0.75 |
| 0.6979 | 1.14 | 550 | 0.5649 | 0.75 |
| 0.6979 | 1.24 | 600 | 0.5628 | 0.75 |
| 0.6979 | 1.34 | 650 | 0.5631 | 0.75 |
| 0.6979 | 1.45 | 700 | 0.5688 | 0.75 |
| 0.6979 | 1.55 | 750 | 0.5624 | 0.75 |
| 0.6979 | 1.65 | 800 | 0.5630 | 0.75 |
| 0.6979 | 1.76 | 850 | 0.5628 | 0.75 |
| 0.6979 | 1.86 | 900 | 0.5624 | 0.75 |
| 0.6979 | 1.96 | 950 | 0.5637 | 0.75 |
| 0.5706 | 2.07 | 1000 | 0.5634 | 0.75 |
Framework versions
- Transformers 4.35.0.dev0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.14.1