bert_sst2_padding70model

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: 0.6816
• Accuracy: 0.9209

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	0.2422	0.9121
0.3225	2.0	866	0.3330	0.9094
0.15	3.0	1299	0.4125	0.9193
0.0796	4.0	1732	0.4849	0.9088
0.033	5.0	2165	0.6146	0.9023
0.0252	6.0	2598	0.5862	0.9105
0.0147	7.0	3031	0.6562	0.9121
0.0147	8.0	3464	0.6735	0.9171
0.01	9.0	3897	0.7122	0.9099
0.017	10.0	4330	0.6584	0.9149
0.0106	11.0	4763	0.7113	0.9171
0.0077	12.0	5196	0.7330	0.9149
0.0108	13.0	5629	0.6942	0.9143
0.0126	14.0	6062	0.6131	0.9160
0.0126	15.0	6495	0.6609	0.9182
0.0074	16.0	6928	0.6579	0.9193
0.0075	17.0	7361	0.6388	0.9220
0.002	18.0	7794	0.6524	0.9253
0.0014	19.0	8227	0.6741	0.9209
0.0009	20.0	8660	0.6816	0.9209

Framework versions

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