|
|
--- |
|
|
library_name: transformers |
|
|
base_model: UBC-NLP/MARBERT |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: questions |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
|
|
# questions |
|
|
|
|
|
This model is a fine-tuned version of [UBC-NLP/MARBERT](https://huggingface.co/UBC-NLP/MARBERT) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.3756 |
|
|
- Micro F1: 0.6542 |
|
|
- Weighted F1: 0.5875 |
|
|
- Jaccard: 0.5726 |
|
|
|
|
|
## 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: 32 |
|
|
- seed: 42 |
|
|
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 5 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Micro F1 | Weighted F1 | Jaccard | |
|
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------:|:-------:| |
|
|
| 0.5395 | 1.0 | 18 | 0.4255 | 0.6375 | 0.5254 | 0.5017 | |
|
|
| 0.4424 | 2.0 | 36 | 0.3918 | 0.6607 | 0.5539 | 0.5743 | |
|
|
| 0.4146 | 3.0 | 54 | 0.3756 | 0.6542 | 0.5875 | 0.5726 | |
|
|
| 0.3844 | 4.0 | 72 | 0.3776 | 0.6197 | 0.5665 | 0.5298 | |
|
|
| 0.3647 | 5.0 | 90 | 0.3738 | 0.6161 | 0.5609 | 0.5286 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.53.3 |
|
|
- Pytorch 2.6.0+cu124 |
|
|
- Datasets 4.0.0 |
|
|
- Tokenizers 0.21.2 |
|
|
|
