|
--- |
|
license: mit |
|
base_model: MilaNLProc/hate-ita-xlm-r-large |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: haspeech_ita |
|
results: [] |
|
language: |
|
- it |
|
datasets: |
|
- Paul/hatecheck-italian |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# haspeech_ita |
|
|
|
This model is a fine-tuned version of [MilaNLProc/hate-ita-xlm-r-large](https://huggingface.co/MilaNLProc/hate-ita-xlm-r-large) |
|
on a dataset [Paul/hatecheck-italian](https://huggingface.co/datasets/Paul/hatecheck-italian). |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0030 |
|
- Accuracy: 0.9973 |
|
|
|
## 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_steps: 500 |
|
- num_epochs: 3 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 0.3538 | 0.48 | 100 | 0.2109 | 0.9241 | |
|
| 0.18 | 0.96 | 200 | 0.1058 | 0.9783 | |
|
| 0.0924 | 1.44 | 300 | 0.0618 | 0.9892 | |
|
| 0.0948 | 1.92 | 400 | 0.0382 | 0.9892 | |
|
| 0.0475 | 2.4 | 500 | 0.0607 | 0.9919 | |
|
| 0.0572 | 2.88 | 600 | 0.0030 | 0.9973 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.0.1 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
