metadata
base_model: vinai/bertweet-base
tags:
- generated_from_trainer
metrics:
- f1
- recall
model-index:
- name: bertweet-base
results: []
bertweet-base
This model is a fine-tuned version of vinai/bertweet-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5690
- F1 Macro: 0.7699
- F1: 0.8459
- F1 Neg: 0.6940
- Acc: 0.795
- Prec: 0.8152
- Recall: 0.8789
- Mcc: 0.5446
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: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 | F1 Neg | Acc | Prec | Recall | Mcc |
|---|---|---|---|---|---|---|---|---|---|---|
| 0.6752 | 1.0 | 592 | 0.6373 | 0.5685 | 0.8134 | 0.3237 | 0.7075 | 0.6873 | 0.9961 | 0.3527 |
| 0.4957 | 2.0 | 1184 | 0.7489 | 0.7577 | 0.8122 | 0.7032 | 0.77 | 0.8504 | 0.7773 | 0.5205 |
| 0.4233 | 3.0 | 1776 | 0.5690 | 0.7699 | 0.8459 | 0.6940 | 0.795 | 0.8152 | 0.8789 | 0.5446 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2