prova_Classi
This model is a fine-tuned version of distilbert-base-uncased on the tweet_eval dataset. It achieves the following results on the evaluation set:
- Loss: 1.5530
- Accuracy: 0.716
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: 0.00013441028267541125
- train_batch_size: 32
- eval_batch_size: 16
- seed: 17
- 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 | Accuracy |
|---|---|---|---|---|
| 0.7022 | 1.0 | 1426 | 0.6581 | 0.7105 |
| 0.5199 | 2.0 | 2852 | 0.6835 | 0.706 |
| 0.2923 | 3.0 | 4278 | 0.7941 | 0.7075 |
| 0.1366 | 4.0 | 5704 | 1.0761 | 0.7115 |
| 0.0645 | 5.0 | 7130 | 1.5530 | 0.716 |
Framework versions
- Transformers 4.13.0
- Pytorch 1.10.0+cu111
- Datasets 1.16.1
- Tokenizers 0.10.3
- Downloads last month
- 14
Dataset used to train marcolatella/tweet_eval_bench
Evaluation results
- Accuracy on tweet_evalself-reported0.716