N_distilbert_sst5_padding50model

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: 4.0506
• Accuracy: 0.5005

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
1.3199	1.0	534	1.3313	0.4163
1.0549	2.0	1068	1.1141	0.5149
0.8561	3.0	1602	1.1560	0.5176
0.6957	4.0	2136	1.3112	0.5041
0.5343	5.0	2670	1.5541	0.5005
0.4112	6.0	3204	1.7335	0.5
0.3204	7.0	3738	1.9688	0.4946
0.2301	8.0	4272	2.1828	0.5054
0.1815	9.0	4806	2.4761	0.4950
0.1475	10.0	5340	2.6397	0.4955
0.1297	11.0	5874	3.0079	0.5032
0.1124	12.0	6408	3.2400	0.4955
0.0762	13.0	6942	3.4782	0.4891
0.0701	14.0	7476	3.5453	0.4982
0.0458	15.0	8010	3.7951	0.4914
0.0367	16.0	8544	3.8300	0.5009
0.0255	17.0	9078	3.9735	0.5027
0.0276	18.0	9612	4.0115	0.5014
0.0312	19.0	10146	4.0131	0.5018
0.018	20.0	10680	4.0506	0.5005

Framework versions

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