my_awesome_model

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

• Loss: 0.0266
• Accuracy: 0.9963
• Recall: 0.8

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: 16
• eval_batch_size: 16
• 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	Accuracy	Recall
No log	1.0	90	0.0396	0.9815	0.0
No log	2.0	180	0.0814	0.9815	0.0
No log	3.0	270	0.0400	0.9914	0.5333
No log	4.0	360	0.8444	0.6510	0.9333
No log	5.0	450	0.0246	0.9951	0.8667
0.0675	6.0	540	0.0686	0.9803	0.9333
0.0675	7.0	630	0.0205	0.9963	0.8667
0.0675	8.0	720	0.0163	0.9975	0.8667
0.0675	9.0	810	0.0183	0.9975	0.8667
0.0675	10.0	900	0.0274	0.9963	0.8
0.0675	11.0	990	0.0309	0.9951	0.7333
0.0107	12.0	1080	0.0268	0.9963	0.8
0.0107	13.0	1170	0.0203	0.9975	0.8667
0.0107	14.0	1260	0.0208	0.9975	0.8667
0.0107	15.0	1350	0.0230	0.9963	0.8
0.0107	16.0	1440	0.0301	0.9963	0.8
0.004	17.0	1530	0.0284	0.9963	0.8
0.004	18.0	1620	0.0282	0.9963	0.8
0.004	19.0	1710	0.0271	0.9963	0.8
0.004	20.0	1800	0.0266	0.9963	0.8

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1