metadata

language:
  - vi
license: mit
base_model: FacebookAI/xlm-roberta-base
tags:
  - generated_from_trainer
model-index:
  - name: xlm-roberta-base_covid_ner_full
    results: []

xlm-roberta-base_covid_ner_full

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the covid19_ner dataset. It achieves the following results on the evaluation set:

Loss: 0.0873
Patient Id: 0.9883
Name: 0.9446
Gender: 0.9785
Age: 0.9765
Job: 0.7063
Location: 0.9538
Organization: 0.8807
Date: 0.9851
Symptom And Disease: 0.8886
Transportation: 1.0
F1 Macro: 0.9302
F1 Micro: 0.9499

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: 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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Patient Id	Name	Gender	Age	Job	Location	Organization	Date	Symptom And Disease	Transportation	F1 Macro	F1 Micro
0.2425	1.0	629	0.0989	0.9661	0.8889	0.8030	0.9068	0.3358	0.9152	0.8045	0.9843	0.8291	0.9222	0.8356	0.9001
0.0596	2.0	1258	0.0885	0.9807	0.9446	0.9283	0.9620	0.4786	0.9462	0.8665	0.9810	0.8690	0.9885	0.8945	0.9367
0.0376	3.0	1887	0.0899	0.9828	0.9284	0.9483	0.9765	0.6406	0.9458	0.8720	0.9865	0.8783	1.0	0.9159	0.9423
0.0257	4.0	2516	0.0919	0.9875	0.9362	0.9766	0.9805	0.6818	0.9505	0.8827	0.9869	0.8871	1.0	0.9270	0.9484
0.0172	5.0	3145	0.0873	0.9883	0.9446	0.9785	0.9765	0.7063	0.9538	0.8807	0.9851	0.8886	1.0	0.9302	0.9499

Framework versions

Transformers 4.41.2
Pytorch 2.1.2
Datasets 2.19.2
Tokenizers 0.19.1