metadata
base_model: salahyahya/weights
tags:
- generated_from_trainer
model-index:
- name: weightsGEC
results: []
weightsGEC
This model is a fine-tuned version of salahyahya/weights on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5494
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 6
- total_train_batch_size: 96
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.6105 | 0.48 | 500 | 0.5669 |
0.6106 | 0.96 | 1000 | 0.5605 |
0.5958 | 1.44 | 1500 | 0.5584 |
0.5927 | 1.92 | 2000 | 0.5549 |
0.5802 | 2.4 | 2500 | 0.5523 |
0.5822 | 2.88 | 3000 | 0.5507 |
0.5758 | 3.36 | 3500 | 0.5514 |
0.5732 | 3.84 | 4000 | 0.5499 |
0.5707 | 4.32 | 4500 | 0.5494 |
0.5716 | 4.8 | 5000 | 0.5494 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.1.2
- Datasets 2.19.2
- Tokenizers 0.19.1