PhoBERT_NER_T4-ABSA

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

• Loss: 0.0135
• F1: 0.9454

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 128
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	435	0.1542	0.6301
No log	2.0	870	0.1053	0.7274
No log	3.0	1305	0.0803	0.7898
No log	4.0	1740	0.0616	0.8217
No log	5.0	2175	0.0504	0.8569
No log	6.0	2610	0.0434	0.8776
No log	7.0	3045	0.0356	0.8870
No log	8.0	3480	0.0308	0.9110
No log	9.0	3915	0.0256	0.9147
No log	10.0	4350	0.0228	0.9241
No log	11.0	4785	0.0208	0.9332
No log	12.0	5220	0.0192	0.9327
No log	13.0	5655	0.0173	0.9408
No log	14.0	6090	0.0161	0.9425
No log	15.0	6525	0.0157	0.9440
No log	16.0	6960	0.0151	0.9391
No log	17.0	7395	0.0142	0.9410
No log	18.0	7830	0.0142	0.9474
No log	19.0	8265	0.0138	0.9463
No log	20.0	8700	0.0135	0.9454

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2