distilbert-base-uncased_fold_5_binary_v1

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: 1.6980
• F1: 0.8110

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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	288	0.4412	0.7981
0.396	2.0	576	0.4419	0.8078
0.396	3.0	864	0.4955	0.8166
0.2019	4.0	1152	0.6341	0.8075
0.2019	5.0	1440	1.0351	0.7979
0.0808	6.0	1728	1.1818	0.7844
0.0315	7.0	2016	1.2530	0.8051
0.0315	8.0	2304	1.3568	0.7937
0.0143	9.0	2592	1.4009	0.8045
0.0143	10.0	2880	1.5333	0.7941
0.0066	11.0	3168	1.5242	0.7982
0.0066	12.0	3456	1.5752	0.8050
0.0091	13.0	3744	1.5199	0.8046
0.0111	14.0	4032	1.5319	0.8117
0.0111	15.0	4320	1.5333	0.8156
0.0072	16.0	4608	1.5461	0.8192
0.0072	17.0	4896	1.5288	0.8252
0.0048	18.0	5184	1.5725	0.8078
0.0048	19.0	5472	1.5896	0.8138
0.0032	20.0	5760	1.6917	0.8071
0.0028	21.0	6048	1.6608	0.8109
0.0028	22.0	6336	1.7013	0.8122
0.0029	23.0	6624	1.6769	0.8148
0.0029	24.0	6912	1.6906	0.8100
0.0006	25.0	7200	1.6980	0.8110

Framework versions

• Transformers 4.21.0
• Pytorch 1.12.0+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1