distilbert-base-uncased-text-classification-v8

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1626
• F1: 0.8539

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	F1
0.435	1.0	1154	0.3097	0.6342
0.2289	2.0	2308	0.1947	0.7906
0.1372	3.0	3462	0.1626	0.8539
0.0867	4.0	4616	0.1693	0.8722
0.0546	5.0	5770	0.1664	0.8796
0.0521	6.0	6924	0.1658	0.8882
0.0447	7.0	8078	0.1651	0.8945
0.0361	8.0	9232	0.1752	0.8922
0.0315	9.0	10386	0.1712	0.8891
0.0289	10.0	11540	0.1763	0.8914
0.0302	11.0	12694	0.1811	0.8937
0.0269	12.0	13848	0.1829	0.8915
0.0334	13.0	15002	0.1867	0.8878
0.0254	14.0	16156	0.1865	0.8869
0.026	15.0	17310	0.1856	0.8936
0.0226	16.0	18464	0.1852	0.8933
0.0226	17.0	19618	0.1843	0.8967
0.0202	18.0	20772	0.1832	0.8984
0.0177	19.0	21926	0.1856	0.8975
0.0186	20.0	23080	0.1854	0.8962

Framework versions

• Transformers 4.38.0
• Pytorch 2.3.1+cu121
• Datasets 2.21.0
• Tokenizers 0.15.2