N_distilbert_sst2_padding40model

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: 0.8411
• Accuracy: 0.9050

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	0.2831	0.8819
0.3503	2.0	866	0.2799	0.8979
0.1786	3.0	1299	0.3919	0.9012
0.087	4.0	1732	0.4669	0.9044
0.063	5.0	2165	0.5653	0.8968
0.023	6.0	2598	0.5849	0.9066
0.0183	7.0	3031	0.6700	0.8984
0.0183	8.0	3464	0.7024	0.9039
0.0107	9.0	3897	0.7215	0.8995
0.0137	10.0	4330	0.7239	0.9072
0.0101	11.0	4763	0.7297	0.9099
0.0094	12.0	5196	0.6947	0.9088
0.0072	13.0	5629	0.6865	0.9110
0.0063	14.0	6062	0.7615	0.9033
0.0063	15.0	6495	0.7822	0.9055
0.0026	16.0	6928	0.8497	0.9017
0.0039	17.0	7361	0.8538	0.8995
0.0031	18.0	7794	0.7964	0.9033
0.0007	19.0	8227	0.8209	0.9050
0.0003	20.0	8660	0.8411	0.9050

Framework versions

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