metadata
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: bertweet-base-finetuned-sentiment-analysis
results: []
bertweet-base-finetuned-sentiment-analysis
This model is a fine-tuned version of cardiffnlp/bertweet-base-sentiment on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.8458
- Accuracy: 0.6426
- F1: 0.6397
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
- 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 | F1 |
|---|---|---|---|---|---|
| 0.8904 | 1.0 | 630 | 0.8509 | 0.6381 | 0.6340 |
| 0.7655 | 2.0 | 1260 | 0.8345 | 0.6579 | 0.6559 |
| 0.66 | 3.0 | 1890 | 0.9199 | 0.6548 | 0.6514 |
| 0.447 | 4.0 | 2520 | 1.0324 | 0.6429 | 0.6417 |
| 0.3585 | 5.0 | 3150 | 1.1234 | 0.6452 | 0.6424 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.10.0+cu111
- Datasets 2.0.0
- Tokenizers 0.12.0