N_distilbert_sst5_padding0model

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: 4.1603
• Accuracy: 0.5100

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
1.2711	1.0	534	1.2114	0.4484
1.0077	2.0	1068	1.0976	0.5303
0.8051	3.0	1602	1.2082	0.5240
0.6379	4.0	2136	1.3809	0.5222
0.4724	5.0	2670	1.6873	0.4932
0.3451	6.0	3204	1.8427	0.5104
0.2629	7.0	3738	2.0872	0.5145
0.1894	8.0	4272	2.3192	0.5163
0.156	9.0	4806	2.6154	0.5136
0.1447	10.0	5340	3.0435	0.4968
0.0956	11.0	5874	3.3029	0.5036
0.084	12.0	6408	3.5400	0.5045
0.0632	13.0	6942	3.7259	0.4995
0.0462	14.0	7476	3.9112	0.4946
0.0378	15.0	8010	4.0038	0.4995
0.0338	16.0	8544	4.0106	0.5118
0.0253	17.0	9078	4.0946	0.5104
0.0299	18.0	9612	4.1451	0.5041
0.022	19.0	10146	4.1501	0.5077
0.0146	20.0	10680	4.1603	0.5100

Framework versions

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