CS505-Classifier-T4_predictLabel_a1_v3

This model is a fine-tuned version of vinai/phobert-base-v2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0095

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: 32
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.98	48	1.0278
No log	1.96	96	0.5582
No log	2.94	144	0.3632
No log	3.92	192	0.2807
No log	4.9	240	0.2237
No log	5.88	288	0.1827
No log	6.86	336	0.1383
No log	7.84	384	0.1124
No log	8.82	432	0.0880
No log	9.8	480	0.0866
0.4209	10.78	528	0.0562
0.4209	11.76	576	0.0444
0.4209	12.73	624	0.0399
0.4209	13.71	672	0.0301
0.4209	14.69	720	0.0262
0.4209	15.67	768	0.0245
0.4209	16.65	816	0.0216
0.4209	17.63	864	0.0207
0.4209	18.61	912	0.0179
0.4209	19.59	960	0.0170
0.0435	20.57	1008	0.0162
0.0435	21.55	1056	0.0130
0.0435	22.53	1104	0.0122
0.0435	23.51	1152	0.0120
0.0435	24.49	1200	0.0107
0.0435	25.47	1248	0.0102
0.0435	26.45	1296	0.0098
0.0435	27.43	1344	0.0096
0.0435	28.41	1392	0.0097
0.0435	29.39	1440	0.0095

Framework versions

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