metadata
license: apache-2.0
base_model: Twitter/twhin-bert-large
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: financial-twhin-bert-large-3labels-tt
results: []
financial-twhin-bert-large-3labels-tt
This model is a fine-tuned version of Twitter/twhin-bert-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2959
- Accuracy: 0.8934
- F1: 0.8943
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: 2.0998212817984933e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.8622 | 0.15 | 100 | 0.6806 | 0.7442 | 0.6936 |
| 0.7005 | 0.3 | 200 | 0.5887 | 0.7731 | 0.7317 |
| 0.5556 | 0.46 | 300 | 0.4472 | 0.8242 | 0.8328 |
| 0.462 | 0.61 | 400 | 0.4188 | 0.8509 | 0.8546 |
| 0.4282 | 0.76 | 500 | 0.3870 | 0.8444 | 0.8498 |
| 0.4109 | 0.91 | 600 | 0.3031 | 0.8818 | 0.8813 |
| 0.3524 | 1.06 | 700 | 0.3483 | 0.8876 | 0.8859 |
| 0.2896 | 1.21 | 800 | 0.3430 | 0.8725 | 0.8757 |
| 0.2731 | 1.37 | 900 | 0.3743 | 0.8602 | 0.8659 |
| 0.2598 | 1.52 | 1000 | 0.3246 | 0.8905 | 0.8917 |
| 0.2954 | 1.67 | 1100 | 0.2988 | 0.8927 | 0.8942 |
| 0.2456 | 1.82 | 1200 | 0.2981 | 0.8912 | 0.8930 |
| 0.2428 | 1.97 | 1300 | 0.2959 | 0.8934 | 0.8943 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2