<!-- 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. -->
platzi-distilroberta-base-mrpc-glue-Jonathan-Castillo
This model is a fine-tuned version of distilroberta-base on the datasetX dataset. It achieves the following results on the evaluation set:
- Loss: 0.6592
- Accuracy: 0.8382
- F1: 0.8850
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
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.5304 | 1.09 | 500 | 0.7888 | 0.7966 | 0.8659 |
| 0.3762 | 2.18 | 1000 | 0.6592 | 0.8382 | 0.8850 |
| 0.2122 | 3.27 | 1500 | 0.9311 | 0.8333 | 0.8828 |
| 0.1345 | 4.36 | 2000 | 0.9803 | 0.8505 | 0.8968 |
| 0.066 | 5.45 | 2500 | 1.0714 | 0.8578 | 0.8968 |
| 0.0306 | 6.54 | 3000 | 1.2510 | 0.8456 | 0.8923 |
| 0.0198 | 7.63 | 3500 | 1.2969 | 0.8456 | 0.8916 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
