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-9H
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: nergrit
type: nergrit
config: nergrit_ner_seacrowd_seq_label
split: test
args: nergrit_ner_seacrowd_seq_label
metrics:
- name: Precision
type: precision
value: 0.9333290962247363
- name: Recall
type: recall
value: 0.9402010371982842
- name: F1
type: f1
value: 0.9367524638790548
- name: Accuracy
type: accuracy
value: 0.9811414616497829
roberta-finetuned-ner-nergrit-9H
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: 0.0982
- Precision: 0.9333
- Recall: 0.9402
- F1: 0.9368
- Accuracy: 0.9811
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 | 0.9995 | 471 | 0.0979 | 0.9354 | 0.9229 | 0.9291 | 0.9795 |
| 0.2005 | 1.9989 | 942 | 0.0967 | 0.9376 | 0.9356 | 0.9366 | 0.9811 |
| 0.0863 | 2.9984 | 1413 | 0.0982 | 0.9333 | 0.9402 | 0.9368 | 0.9811 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1