N_distilbert_sst2_padding30model

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: 0.8867
• Accuracy: 0.9023

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	0.2836	0.8814
0.3476	2.0	866	0.2705	0.9012
0.1836	3.0	1299	0.3451	0.9083
0.0887	4.0	1732	0.4852	0.9066
0.0451	5.0	2165	0.5730	0.9044
0.0245	6.0	2598	0.7197	0.8924
0.0186	7.0	3031	0.6648	0.8990
0.0186	8.0	3464	0.6407	0.9023
0.017	9.0	3897	0.8361	0.8913
0.009	10.0	4330	0.7010	0.9044
0.0169	11.0	4763	0.7497	0.9050
0.0087	12.0	5196	0.7683	0.9039
0.0073	13.0	5629	0.8405	0.8979
0.0036	14.0	6062	0.7964	0.9066
0.0036	15.0	6495	0.8325	0.9055
0.002	16.0	6928	0.8294	0.9039
0.0045	17.0	7361	0.8773	0.8995
0.0019	18.0	7794	0.8825	0.9028
0.0032	19.0	8227	0.9006	0.9023
0.0008	20.0	8660	0.8867	0.9023

Framework versions

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