distilbert_sst2_padding20model

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

• Loss: 0.8375
• Accuracy: 0.9110

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
No log	1.0	433	0.2668	0.8935
0.3473	2.0	866	0.3801	0.8814
0.1828	3.0	1299	0.3816	0.9061
0.0818	4.0	1732	0.5191	0.8973
0.0522	5.0	2165	0.6669	0.8946
0.0251	6.0	2598	0.6408	0.9072
0.0123	7.0	3031	0.6982	0.9006
0.0123	8.0	3464	0.7690	0.8962
0.011	9.0	3897	0.7984	0.9012
0.0057	10.0	4330	0.7417	0.9044
0.02	11.0	4763	0.7771	0.9088
0.0088	12.0	5196	0.7492	0.9033
0.0105	13.0	5629	0.8034	0.9033
0.0077	14.0	6062	0.7967	0.9033
0.0077	15.0	6495	0.8108	0.9088
0.0031	16.0	6928	0.8189	0.9088
0.0033	17.0	7361	0.8219	0.9116
0.0017	18.0	7794	0.8460	0.9077
0.0009	19.0	8227	0.8520	0.9099
0.0002	20.0	8660	0.8375	0.9110

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3