metadata
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: XLMRoberta-base-amazon-massive-Intent
results: []
widget:
- text: staubsauge den flur
datasets:
- AmazonScience/massive
language:
- en
- ru
XLMRoberta-base-amazon-massive-Intent
This model is a fine-tuned version of xlm-roberta-base on the MASSIVE dataset. It achieves the following results on the evaluation set:
- Loss: 0.5620
- Accuracy: 0.8751
- F1: 0.8269
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: 7e-06
- 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
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 2.4641 | 1.0 | 1440 | 1.4258 | 0.6709 | 0.4126 |
| 1.1447 | 2.0 | 2880 | 0.8477 | 0.8060 | 0.6318 |
| 0.7437 | 3.0 | 4320 | 0.6688 | 0.8409 | 0.7060 |
| 0.5543 | 4.0 | 5760 | 0.6006 | 0.8601 | 0.7813 |
| 0.4375 | 5.0 | 7200 | 0.5780 | 0.8635 | 0.7937 |
| 0.3763 | 6.0 | 8640 | 0.5748 | 0.8694 | 0.8170 |
| 0.3265 | 7.0 | 10080 | 0.5620 | 0.8751 | 0.8269 |
| 0.2916 | 8.0 | 11520 | 0.5701 | 0.8756 | 0.8260 |
| 0.2628 | 9.0 | 12960 | 0.5728 | 0.8760 | 0.8271 |
| 0.2474 | 10.0 | 14400 | 0.5740 | 0.8770 | 0.8288 |
Framework versions
- Transformers 4.41.0
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1