N_distilbert_sst5_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: 4.2427
• Accuracy: 0.4919

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
1.3042	1.0	534	1.2788	0.4154
1.0406	2.0	1068	1.1166	0.5127
0.8517	3.0	1602	1.2093	0.5167
0.6903	4.0	2136	1.2954	0.5100
0.5381	5.0	2670	1.5639	0.4959
0.4124	6.0	3204	1.7632	0.4873
0.3138	7.0	3738	1.9780	0.4891
0.225	8.0	4272	2.2458	0.4860
0.1917	9.0	4806	2.4285	0.4950
0.1531	10.0	5340	2.7969	0.4828
0.1247	11.0	5874	3.1633	0.4891
0.0965	12.0	6408	3.3317	0.5023
0.0818	13.0	6942	3.6110	0.4973
0.0746	14.0	7476	3.7011	0.4932
0.044	15.0	8010	3.8336	0.4941
0.036	16.0	8544	4.0305	0.4941
0.0357	17.0	9078	4.1497	0.4928
0.0283	18.0	9612	4.1689	0.4982
0.0214	19.0	10146	4.1801	0.4959
0.0191	20.0	10680	4.2427	0.4919

Framework versions

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