N_distilbert_sst2_padding0model

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: 0.8371
• Accuracy: 0.9072

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	0.2637	0.8951
0.3398	2.0	866	0.2966	0.8979
0.1677	3.0	1299	0.5400	0.8852
0.0745	4.0	1732	0.5148	0.9061
0.0376	5.0	2165	0.5925	0.9017
0.0208	6.0	2598	0.6006	0.9039
0.0179	7.0	3031	0.6727	0.9028
0.0179	8.0	3464	0.6872	0.9050
0.0154	9.0	3897	0.7194	0.9061
0.009	10.0	4330	0.7977	0.8984
0.004	11.0	4763	0.8141	0.9033
0.005	12.0	5196	0.8024	0.9077
0.0048	13.0	5629	0.7693	0.9055
0.0076	14.0	6062	0.7940	0.9094
0.0076	15.0	6495	0.8173	0.9105
0.0063	16.0	6928	0.7915	0.9055
0.0069	17.0	7361	0.8337	0.9033
0.002	18.0	7794	0.8250	0.9077
0.0029	19.0	8227	0.8175	0.9072
0.0011	20.0	8660	0.8371	0.9072

Framework versions

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