vihealthbert-w_dual-ViNLI

This model is a fine-tuned version of demdecuong/vihealthbert-base-word on the tmnam20/ViNLI dataset. It achieves the following results on the evaluation set:

• Loss: 2.6042
• Accuracy: 0.5919

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 30000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
5.8126	15.625	1000	3.5461	0.4450
2.605	31.25	2000	2.7789	0.5404
1.5924	46.875	3000	2.5432	0.5809
1.2233	62.5	4000	2.6662	0.5567
0.9236	78.125	5000	2.4691	0.5927
0.7193	93.75	6000	2.4053	0.6027
0.6259	109.375	7000	2.5938	0.5782
0.5082	125.0	8000	2.4809	0.6137
0.4438	140.625	9000	2.7056	0.5819
0.4075	156.25	10000	2.6501	0.5946
0.3571	171.875	11000	2.5337	0.6082

Framework versions

• Transformers 4.40.2
• Pytorch 2.0.1+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1