metadata
license: mit
base_model: BigTMiami/dapt_plus_tapt_helpfulness_base_pretraining_model
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: dapt_plus_tapt_amazon_helpfulness_classification_v2
results: []
dapt_plus_tapt_amazon_helpfulness_classification_v2
This model is a fine-tuned version of BigTMiami/dapt_plus_tapt_helpfulness_base_pretraining_model on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3267
- Accuracy: 0.8628
- F1 Macro: 0.6898
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=(0.9,0.98) and epsilon=1e-06
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro |
|---|---|---|---|---|---|
| 0.3333 | 1.0 | 1563 | 0.3267 | 0.8628 | 0.6898 |
| 0.309 | 2.0 | 3126 | 0.3859 | 0.8696 | 0.6292 |
| 0.2443 | 3.0 | 4689 | 0.4222 | 0.8664 | 0.6222 |
| 0.1683 | 4.0 | 6252 | 0.5618 | 0.867 | 0.6315 |
| 0.1591 | 5.0 | 7815 | 0.6760 | 0.8668 | 0.6678 |
| 0.0877 | 6.0 | 9378 | 0.8763 | 0.8614 | 0.6678 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2