metadata
license: mit
base_model: FacebookAI/roberta-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: roberta_base_twitter
results: []
roberta_base_twitter
This model is a fine-tuned version of FacebookAI/roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4759
- Accuracy: 0.7711
- F1 Macro: 0.7372
- F1 Micro: 0.7711
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: 32
- eval_batch_size: 32
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 64
- total_eval_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro |
|---|---|---|---|---|---|---|
| 0.4867 | 0.37 | 50 | 0.4759 | 0.7711 | 0.7372 | 0.7711 |
| 0.4633 | 0.74 | 100 | 0.4788 | 0.7711 | 0.7285 | 0.7711 |
| 0.4582 | 1.1 | 150 | 0.4821 | 0.7739 | 0.7356 | 0.7739 |
| 0.4642 | 1.47 | 200 | 0.4841 | 0.7592 | 0.7292 | 0.7592 |
| 0.458 | 1.84 | 250 | 0.4864 | 0.7739 | 0.7369 | 0.7739 |
| 0.4001 | 2.21 | 300 | 0.4867 | 0.7684 | 0.7346 | 0.7684 |
| 0.443 | 2.57 | 350 | 0.4886 | 0.7601 | 0.7258 | 0.7601 |
| 0.3461 | 2.94 | 400 | 0.4942 | 0.7656 | 0.7296 | 0.7656 |
Framework versions
- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2