metadata

license: mit
base_model: xlm-roberta-base
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: ner_model
    results: []
datasets:
  - pythainlp/thainer-corpus-v2
language:
  - th

ner_model

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

Loss: 0.1247
Precision: 0.8073
Recall: 0.8695
F1: 0.8372
Accuracy: 0.9655

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: 1e-05
train_batch_size: 16
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	Precision	Recall	F1	Accuracy
No log	0.4	100	0.5360	0.4604	0.4644	0.4624	0.8846
No log	0.81	200	0.2882	0.6137	0.6619	0.6369	0.9307
No log	1.21	300	0.2128	0.7236	0.7649	0.7437	0.9442
No log	1.62	400	0.1811	0.7146	0.7925	0.7515	0.9494
0.4608	2.02	500	0.1594	0.7369	0.8021	0.7681	0.9542
0.4608	2.43	600	0.1532	0.7494	0.8331	0.7890	0.9572
0.4608	2.83	700	0.1403	0.7660	0.8417	0.8021	0.9594
0.4608	3.24	800	0.1342	0.7909	0.8428	0.8160	0.9625
0.4608	3.64	900	0.1325	0.7867	0.8572	0.8204	0.9626
0.1256	4.05	1000	0.1275	0.8056	0.8632	0.8334	0.9648
0.1256	4.45	1100	0.1229	0.8131	0.8643	0.8379	0.9657
0.1256	4.86	1200	0.1247	0.8073	0.8695	0.8372	0.9655

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.0