bert_sst2_padding30model

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: 0.8153
• Accuracy: 0.9116

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	0.2372	0.9099
0.3252	2.0	866	0.2798	0.9138
0.1573	3.0	1299	0.4393	0.9033
0.0784	4.0	1732	0.4861	0.9171
0.0317	5.0	2165	0.6050	0.9099
0.0163	6.0	2598	0.6197	0.9143
0.0197	7.0	3031	0.7356	0.9050
0.0197	8.0	3464	0.6162	0.9143
0.0179	9.0	3897	0.6469	0.9176
0.0082	10.0	4330	0.7268	0.9121
0.0077	11.0	4763	0.7143	0.9127
0.0079	12.0	5196	0.6712	0.9105
0.0092	13.0	5629	0.7329	0.9165
0.0021	14.0	6062	0.8041	0.9094
0.0021	15.0	6495	0.7865	0.9138
0.0035	16.0	6928	0.8221	0.9121
0.0029	17.0	7361	0.8066	0.9132
0.0035	18.0	7794	0.7961	0.9149
0.0002	19.0	8227	0.8223	0.9138
0.0	20.0	8660	0.8153	0.9116

Framework versions

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