bert_sst2_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: 0.7522
• Accuracy: 0.9182

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	0.2291	0.9050
0.3211	2.0	866	0.2961	0.9099
0.1454	3.0	1299	0.3994	0.9187
0.0691	4.0	1732	0.4780	0.9187
0.0313	5.0	2165	0.5999	0.9072
0.0221	6.0	2598	0.6512	0.9083
0.0107	7.0	3031	0.7013	0.9138
0.0107	8.0	3464	0.7411	0.9121
0.01	9.0	3897	0.5985	0.9127
0.0145	10.0	4330	0.6901	0.9127
0.0083	11.0	4763	0.6588	0.9198
0.0111	12.0	5196	0.6871	0.9127
0.0062	13.0	5629	0.7503	0.9132
0.0095	14.0	6062	0.7295	0.9160
0.0095	15.0	6495	0.7709	0.9149
0.0019	16.0	6928	0.7023	0.9138
0.0037	17.0	7361	0.7238	0.9165
0.004	18.0	7794	0.7378	0.9160
0.0003	19.0	8227	0.7574	0.9193
0.0	20.0	8660	0.7522	0.9182

Framework versions

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