bert_sst2_padding90model

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: 0.8168
• Accuracy: 0.9143

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	0.2313	0.9077
0.3194	2.0	866	0.3335	0.9149
0.1491	3.0	1299	0.3656	0.9127
0.0757	4.0	1732	0.4625	0.9171
0.0373	5.0	2165	0.8161	0.8808
0.0259	6.0	2598	0.5827	0.9193
0.0112	7.0	3031	0.7514	0.9061
0.0112	8.0	3464	0.6847	0.9182
0.0109	9.0	3897	0.5902	0.9176
0.0143	10.0	4330	0.6670	0.9154
0.0104	11.0	4763	0.6928	0.9154
0.0058	12.0	5196	0.7326	0.9149
0.0059	13.0	5629	0.7084	0.9154
0.0066	14.0	6062	0.6887	0.9204
0.0066	15.0	6495	0.6676	0.9226
0.0073	16.0	6928	0.7323	0.9176
0.0044	17.0	7361	0.7690	0.9171
0.0014	18.0	7794	0.7720	0.9187
0.0001	19.0	8227	0.7994	0.9154
0.0017	20.0	8660	0.8168	0.9143

Framework versions

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