metadata

base_model: vinai/phobert-base-v2
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - recall
  - precision
model-index:
  - name: cls-comment-phobert-base-v2-v2.4
    results: []

cls-comment-phobert-base-v2-v2.4

This model is a fine-tuned version of vinai/phobert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.6177
Accuracy: 0.9241
F1 Score: 0.8919
Recall: 0.8930
Precision: 0.8911

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: 64
eval_batch_size: 64
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 128
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
training_steps: 4000
label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Score	Recall	Precision
1.7008	0.96	100	1.5259	0.4623	0.1091	0.1687	0.2296
1.4089	1.91	200	1.1875	0.6568	0.2499	0.2948	0.2170
1.0776	2.87	300	0.9009	0.8038	0.5304	0.5309	0.5333
0.8625	3.83	400	0.7617	0.8575	0.6321	0.6372	0.7107
0.7245	4.78	500	0.6894	0.8818	0.7282	0.7214	0.8803
0.6573	5.74	600	0.6651	0.8968	0.8406	0.8213	0.8770
0.6082	6.7	700	0.6335	0.9079	0.8630	0.8667	0.8595
0.5674	7.66	800	0.6363	0.9106	0.8692	0.8795	0.8621
0.5477	8.61	900	0.6269	0.9151	0.8776	0.8693	0.8877
0.5256	9.57	1000	0.6178	0.9205	0.8835	0.8849	0.8826
0.5148	10.53	1100	0.6214	0.9199	0.8796	0.8839	0.8762
0.4999	11.48	1200	0.6158	0.9229	0.8856	0.8853	0.8862
0.4916	12.44	1300	0.6186	0.9232	0.8839	0.8795	0.8888
0.479	13.4	1400	0.6285	0.9202	0.8847	0.8833	0.8864
0.4812	14.35	1500	0.6177	0.9241	0.8919	0.8930	0.8911
0.4667	15.31	1600	0.6206	0.9256	0.8848	0.8853	0.8843
0.4668	16.27	1700	0.6201	0.9265	0.8854	0.8876	0.8837
0.4635	17.22	1800	0.6252	0.9253	0.8901	0.8877	0.8927
0.4593	18.18	1900	0.6264	0.9274	0.8891	0.8899	0.8887
0.4538	19.14	2000	0.6228	0.9265	0.8891	0.8913	0.8870

Framework versions

Transformers 4.38.2
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.15.2