|
|
--- |
|
|
license: mit |
|
|
base_model: xlm-roberta-large |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: xlm_r_large-baseline_model-v2-fallen-oath-3 |
|
|
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. --> |
|
|
|
|
|
# xlm_r_large-baseline_model-v2-fallen-oath-3 |
|
|
|
|
|
This model is a fine-tuned version of [xlm-roberta-large](https://huggingface.co/xlm-roberta-large) on [SOLD](https://huggingface.co/datasets/sinhala-nlp/SOLD) dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.5036 |
|
|
- Precision 0: 0.8766 |
|
|
- Precision 1: 0.7911 |
|
|
- Recall 0: 0.8512 |
|
|
- Recall 1: 0.8246 |
|
|
- F1 0: 0.8637 |
|
|
- F1 1: 0.8075 |
|
|
- Precision Weighted: 0.8419 |
|
|
- Recall Weighted: 0.8404 |
|
|
- F1 Weighted: 0.8409 |
|
|
- F1 Macro: 0.8356 |
|
|
|
|
|
## 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 |
|
|
- lr_scheduler_warmup_ratio: 0.1 |
|
|
- num_epochs: 5 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision 0 | Precision 1 | Recall 0 | Recall 1 | F1 0 | F1 1 | Precision Weighted | Recall Weighted | F1 Weighted | F1 Macro | |
|
|
|:-------------:|:-----:|:----:|:---------------:|:-----------:|:-----------:|:--------:|:--------:|:------:|:------:|:------------------:|:---------------:|:-----------:|:--------:| |
|
|
| 0.4937 | 1.0 | 469 | 0.4268 | 0.8346 | 0.7933 | 0.8667 | 0.7488 | 0.8503 | 0.7704 | 0.8179 | 0.8188 | 0.8179 | 0.8104 | |
|
|
| 0.3945 | 2.0 | 938 | 0.3987 | 0.9083 | 0.7168 | 0.7603 | 0.8877 | 0.8277 | 0.7931 | 0.8305 | 0.812 | 0.8137 | 0.8104 | |
|
|
| 0.3721 | 3.0 | 1407 | 0.3612 | 0.8654 | 0.7992 | 0.8620 | 0.8039 | 0.8637 | 0.8016 | 0.8386 | 0.8384 | 0.8385 | 0.8326 | |
|
|
| 0.2721 | 4.0 | 1876 | 0.4191 | 0.8514 | 0.8246 | 0.8875 | 0.7734 | 0.8691 | 0.7982 | 0.8405 | 0.8412 | 0.8403 | 0.8336 | |
|
|
| 0.2144 | 5.0 | 2345 | 0.5036 | 0.8766 | 0.7911 | 0.8512 | 0.8246 | 0.8637 | 0.8075 | 0.8419 | 0.8404 | 0.8409 | 0.8356 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.41.2 |
|
|
- Pytorch 2.3.0+cu121 |
|
|
- Datasets 2.19.2 |
|
|
- Tokenizers 0.19.1 |
|
|
|
