N_distilbert_sst2_padding10model

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

• Loss: 0.8293
• Accuracy: 0.9083

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	0.2672	0.8929
0.3404	2.0	866	0.2730	0.9072
0.18	3.0	1299	0.3929	0.8979
0.0942	4.0	1732	0.5054	0.9006
0.048	5.0	2165	0.6373	0.8885
0.0215	6.0	2598	0.5832	0.9055
0.0115	7.0	3031	0.6447	0.9061
0.0115	8.0	3464	0.7011	0.8995
0.0121	9.0	3897	0.7374	0.9006
0.0116	10.0	4330	0.6765	0.9072
0.0133	11.0	4763	0.7147	0.9028
0.0091	12.0	5196	0.7716	0.9072
0.0043	13.0	5629	0.7582	0.9072
0.0082	14.0	6062	0.8606	0.8951
0.0082	15.0	6495	0.7580	0.9094
0.0032	16.0	6928	0.8095	0.9083
0.0024	17.0	7361	0.8023	0.9099
0.0024	18.0	7794	0.8096	0.9088
0.0038	19.0	8227	0.8086	0.9099
0.001	20.0	8660	0.8293	0.9083

Framework versions

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