metadata
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: XLMRoberta-base-amazon-massive-NER
results: []
widget:
- text: Maria has an exam at five am this week
datasets:
- AmazonScience/massive
language:
- en
- ru
XLMRoberta-base-amazon-massive-NER
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: 0.2907
- Precision: 0.6189
- Recall: 0.6243
- F1: 0.6123
- Accuracy: 0.9200
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: 32
- eval_batch_size: 32
- 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 | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.9645 | 1.0 | 720 | 0.4148 | 0.4631 | 0.4177 | 0.4154 | 0.8950 |
| 0.4421 | 2.0 | 1440 | 0.3181 | 0.5808 | 0.6001 | 0.5780 | 0.9154 |
| 0.2514 | 3.0 | 2160 | 0.2907 | 0.6189 | 0.6243 | 0.6123 | 0.9200 |
| 0.2117 | 4.0 | 2880 | 0.2967 | 0.6522 | 0.6351 | 0.6352 | 0.9252 |
| 0.1592 | 5.0 | 3600 | 0.3090 | 0.6288 | 0.6923 | 0.6520 | 0.9233 |
| 0.131 | 6.0 | 4320 | 0.2961 | 0.6619 | 0.6693 | 0.6546 | 0.9282 |
| 0.1054 | 7.0 | 5040 | 0.3147 | 0.6424 | 0.6762 | 0.6498 | 0.9260 |
| 0.0923 | 8.0 | 5760 | 0.3171 | 0.6447 | 0.6945 | 0.6614 | 0.9257 |
| 0.0845 | 9.0 | 6480 | 0.3328 | 0.6434 | 0.6791 | 0.6539 | 0.9256 |
| 0.0691 | 10.0 | 7200 | 0.3314 | 0.6628 | 0.6834 | 0.6635 | 0.9264 |
Framework versions
- Transformers 4.41.0
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1