metadata
license: mit
base_model: FacebookAI/xlm-roberta-large
tags:
- generated_from_trainer
datasets:
- nergrit
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: roberta-finetuned-ner-nergrit-8H-light
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: nergrit
type: nergrit
config: nergrit_ner_seacrowd_seq_label
split: validation
args: nergrit_ner_seacrowd_seq_label
metrics:
- name: Precision
type: precision
value: 0.981006671007531
- name: Recall
type: recall
value: 0.9810548818694482
- name: F1
type: f1
value: 0.9810307758461823
- name: Accuracy
type: accuracy
value: 0.9772770466099682
roberta-finetuned-ner-nergrit-8H-light
This model is a fine-tuned version of FacebookAI/xlm-roberta-large on the nergrit dataset. It achieves the following results on the evaluation set:
- Loss: 0.1130
- Precision: 0.9810
- Recall: 0.9811
- F1: 0.9810
- Accuracy: 0.9773
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: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 0.9994 | 392 | 0.1196 | 0.9793 | 0.9800 | 0.9796 | 0.9757 |
| 0.1919 | 1.9987 | 784 | 0.1048 | 0.9810 | 0.9814 | 0.9812 | 0.9775 |
| 0.0823 | 2.9981 | 1176 | 0.1130 | 0.9810 | 0.9811 | 0.9810 | 0.9773 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1