roberta-large-mnli for title-genre classification
This model is a fine-tuned version of roberta-large-mnli on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2758
- F1: 0.5464
Model description
This classifies one or more genre labels in a multi-label setting for a given book title.
The 'standard' way of interpreting the predictions is that the predicted labels for a given example are only the ones with a greater than 50% probability.
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 6e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-10
- lr_scheduler_type: linear
- num_epochs: 5.0
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.3096 | 1.0 | 62 | 0.2862 | 0.3707 |
| 0.2863 | 2.0 | 124 | 0.2804 | 0.4422 |
| 0.2618 | 3.0 | 186 | 0.2773 | 0.4989 |
| 0.2432 | 4.0 | 248 | 0.2764 | 0.5223 |
| 0.2241 | 5.0 | 310 | 0.2758 | 0.5464 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.2.0.dev20231001+cu121
- Datasets 2.14.5
- Tokenizers 0.13.3
