N_bert_sst5_padding40model

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

• Loss: 4.2139
• Accuracy: 0.5344

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
1.2735	1.0	534	1.2001	0.4701
0.9729	2.0	1068	1.0735	0.5367
0.7608	3.0	1602	1.1589	0.5258
0.5806	4.0	2136	1.4040	0.5226
0.4182	5.0	2670	1.6600	0.5158
0.3021	6.0	3204	1.9966	0.5072
0.2286	7.0	3738	2.2465	0.5271
0.17	8.0	4272	2.5603	0.5258
0.1489	9.0	4806	2.9361	0.5086
0.1191	10.0	5340	3.1244	0.5208
0.0966	11.0	5874	3.4286	0.5195
0.0678	12.0	6408	3.6056	0.5195
0.0762	13.0	6942	3.6478	0.5376
0.0537	14.0	7476	3.8186	0.5258
0.0269	15.0	8010	4.0681	0.5204
0.0272	16.0	8544	4.0600	0.5380
0.0168	17.0	9078	4.1078	0.5285
0.0143	18.0	9612	4.1354	0.5371
0.0138	19.0	10146	4.2105	0.5344
0.0114	20.0	10680	4.2139	0.5344

Framework versions

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