CS505-NerCOQE-PhoBERT-Predicate

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: 0.0001
• F1: 1.0

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	53	0.2653	0.3142
No log	2.0	106	0.1731	0.5180
No log	3.0	159	0.0786	0.6748
No log	4.0	212	0.0437	0.8107
No log	5.0	265	0.0220	0.8908
No log	6.0	318	0.0159	0.9303
No log	7.0	371	0.0080	0.9734
No log	8.0	424	0.0050	0.9693
No log	9.0	477	0.0029	0.9876
No log	10.0	530	0.0011	0.9976
No log	11.0	583	0.0017	0.9929
No log	12.0	636	0.0019	0.9917
No log	13.0	689	0.0008	0.9976
No log	14.0	742	0.0005	0.9988
No log	15.0	795	0.0002	0.9988
No log	16.0	848	0.0001	1.0
No log	17.0	901	0.0001	1.0
No log	18.0	954	0.0001	1.0
No log	19.0	1007	0.0001	1.0
No log	20.0	1060	0.0001	1.0

Framework versions

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