CS505-Classifier-T4_predictLabel_a1_v2

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

• Loss: 0.0077

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: 25

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.98	48	1.0151
No log	1.96	96	0.5423
No log	2.94	144	0.3287
No log	3.92	192	0.2296
No log	4.9	240	0.1795
No log	5.88	288	0.1419
No log	6.86	336	0.1083
No log	7.84	384	0.0807
No log	8.82	432	0.0609
No log	9.8	480	0.0614
0.3965	10.78	528	0.0349
0.3965	11.76	576	0.0289
0.3965	12.73	624	0.0252
0.3965	13.71	672	0.0193
0.3965	14.69	720	0.0163
0.3965	15.67	768	0.0147
0.3965	16.65	816	0.0139
0.3965	17.63	864	0.0134
0.3965	18.61	912	0.0114
0.3965	19.59	960	0.0100
0.0339	20.57	1008	0.0083
0.0339	21.55	1056	0.0079
0.0339	22.53	1104	0.0077
0.0339	23.51	1152	0.0081
0.0339	24.49	1200	0.0077

Framework versions

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