bert-base-uncased-finetuned-clinc_oos-distilled-clinc_oos

This model is a fine-tuned version of distilbert-base-uncased on the clinc_oos dataset. It achieves the following results on the evaluation set:

• Loss: 0.7724
• Accuracy: 0.9158

Model Training Details

Parameter	Value
Task	text-classification
Teacher Model	bert-base-uncased-finetuned-clinc_oos
Student Model	distilbert-base-uncased
Dataset Name	clinc_oos
Dataset Config	plus
Evaluation Dataset	validation
Batch Size	48
Number of Epochs	5
Learning Rate	0.00002
Alpha*	1
*alpha: (Total_loss = alpha * Loss_CE + (1-alpha) * Loss_KD)

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: 48
• eval_batch_size: 48
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	318	3.2762	0.7284
3.7824	2.0	636	1.8624	0.8358
3.7824	3.0	954	1.1512	0.8984
1.6858	4.0	1272	0.8540	0.9132
0.8983	5.0	1590	0.7724	0.9158

Framework versions

• Transformers 4.33.3
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3