metadata
license: mit
base_model: microsoft/deberta-v3-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: sentiment-analysis-whatsapp
results: []
sentiment-analysis-whatsapp
This model is a fine-tuned version of microsoft/deberta-v3-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2229
- Accuracy: {'accuracy': 0.929}
- F1 Macro: 0.9285
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: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 99
- gradient_accumulation_steps: 5
- total_train_batch_size: 320
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro |
|---|---|---|---|---|---|
| No log | 1.0 | 50 | 0.6396 | {'accuracy': 0.7845} | 0.7828 |
| No log | 2.0 | 100 | 0.2665 | {'accuracy': 0.915} | 0.9145 |
| No log | 3.0 | 150 | 0.2229 | {'accuracy': 0.929} | 0.9285 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2