distilbert-base-uncased-distilled-squad-finetuned-clinc

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

• Loss: 1.7920
• Accuracy: 0.8723

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: 384
• eval_batch_size: 384
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	40	3.7816	0.2016
No log	2.0	80	3.3589	0.5374
No log	3.0	120	2.9695	0.6955
No log	4.0	160	2.6408	0.7726
No log	5.0	200	2.3697	0.8145
No log	6.0	240	2.1547	0.8426
No log	7.0	280	1.9912	0.8529
2.8639	8.0	320	1.8802	0.8645
2.8639	9.0	360	1.8138	0.8706
2.8639	10.0	400	1.7920	0.8723

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0
• Datasets 2.14.6
• Tokenizers 0.13.3