metadata
license: mit
tags:
- generated_from_trainer
metrics:
- f1
- recall
- precision
model-index:
- name: sentiment-deberta-e6-b16
results: []
sentiment-deberta-e6-b16
This model is a fine-tuned version of Elron/deberta-v3-large-sentiment on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.1021
- F1: 0.6226
- Recall: 0.6226
- Precision: 0.6226
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: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Recall | Precision |
|---|---|---|---|---|---|---|
| No log | 1.0 | 375 | 0.8565 | 0.6792 | 0.6792 | 0.6792 |
| 0.871 | 2.0 | 750 | 1.0402 | 0.5930 | 0.5930 | 0.5930 |
| 0.4299 | 3.0 | 1125 | 1.3899 | 0.5876 | 0.5876 | 0.5876 |
| 0.1941 | 4.0 | 1500 | 1.5427 | 0.6307 | 0.6307 | 0.6307 |
| 0.1941 | 5.0 | 1875 | 1.9280 | 0.6469 | 0.6469 | 0.6469 |
| 0.0802 | 6.0 | 2250 | 2.1021 | 0.6226 | 0.6226 | 0.6226 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3