cls-comment-phobert-base-v2-v2.1

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: 0.4105
• Accuracy: 0.8994
• F1 Score: 0.8400
• Recall: 0.8281
• Precision: 0.8593

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: 1500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Score	Recall	Precision
1.5954	2.02	100	1.2503	0.6325	0.2395	0.2751	0.2173
0.9847	4.04	200	0.6986	0.8006	0.5335	0.5327	0.5376
0.6015	6.06	300	0.5355	0.8503	0.6428	0.6491	0.6730
0.4411	8.08	400	0.4605	0.8701	0.6817	0.6700	0.7111
0.3344	10.1	500	0.4352	0.8777	0.6919	0.6766	0.7176
0.2843	12.12	600	0.4262	0.8796	0.6978	0.6900	0.7078
0.2288	14.14	700	0.4166	0.8745	0.6927	0.6868	0.7100
0.1816	16.16	800	0.4045	0.8879	0.8109	0.7761	0.8668
0.1597	18.18	900	0.4193	0.8841	0.8206	0.7973	0.8523
0.1306	20.2	1000	0.3999	0.8955	0.8349	0.8059	0.8747
0.1182	22.22	1100	0.4060	0.8936	0.8313	0.8057	0.8672
0.1046	24.24	1200	0.4133	0.8955	0.8351	0.8312	0.8460
0.0973	26.26	1300	0.4145	0.8955	0.8319	0.8142	0.8590
0.088	28.28	1400	0.4096	0.8994	0.8412	0.8283	0.8612
0.0893	30.3	1500	0.4105	0.8994	0.8400	0.8281	0.8593

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2