metadata
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: xlm-roberta-base-finetuned-ner-thesis-dseb
results: []
xlm-roberta-base-finetuned-ner-thesis-dseb
This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1007
- Precision: 0.5789
- Recall: 0.7857
- F1: 0.6667
- Accuracy: 0.9871
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.0117 | 1.0 | 50 | 0.2576 | 0.3256 | 0.35 | 0.3373 | 0.9523 |
0.0454 | 2.0 | 100 | 0.2768 | 0.5714 | 0.3 | 0.3934 | 0.9590 |
0.0188 | 3.0 | 150 | 0.1758 | 0.6429 | 0.45 | 0.5294 | 0.9746 |
0.0144 | 4.0 | 200 | 0.3266 | 0.5714 | 0.2 | 0.2963 | 0.9601 |
0.0134 | 5.0 | 250 | 0.2405 | 0.7143 | 0.375 | 0.4918 | 0.9667 |
0.0038 | 6.0 | 300 | 0.1727 | 0.5660 | 0.75 | 0.6452 | 0.9759 |
0.0036 | 7.0 | 350 | 0.1335 | 0.7561 | 0.775 | 0.7654 | 0.9835 |
0.0047 | 8.0 | 400 | 0.1240 | 0.7111 | 0.8 | 0.7529 | 0.9836 |
0.0013 | 9.0 | 450 | 0.1468 | 0.8 | 0.7 | 0.7467 | 0.9782 |
0.0001 | 10.0 | 500 | 0.1222 | 0.7368 | 0.7 | 0.7179 | 0.9811 |
0.0 | 11.0 | 550 | 0.1261 | 0.7368 | 0.7 | 0.7179 | 0.9817 |
0.0 | 12.0 | 600 | 0.1273 | 0.7368 | 0.7 | 0.7179 | 0.9817 |
0.0 | 13.0 | 650 | 0.1293 | 0.7368 | 0.7 | 0.7179 | 0.9809 |
0.0001 | 14.0 | 700 | 0.1367 | 0.7838 | 0.725 | 0.7532 | 0.9809 |
0.0003 | 15.0 | 750 | 0.1383 | 0.8056 | 0.725 | 0.7632 | 0.9808 |
Framework versions
- Transformers 4.42.3
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1