distilbert-base-uncased_fold_7_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: 1.8361
• F1: 0.7958

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	0.4025	0.8071
0.3986	2.0	576	0.3979	0.8072
0.3986	3.0	864	0.5170	0.8041
0.1761	4.0	1152	0.7946	0.7940
0.1761	5.0	1440	1.0000	0.7937
0.0705	6.0	1728	1.1484	0.7875
0.0294	7.0	2016	1.1548	0.8042
0.0294	8.0	2304	1.3036	0.8069
0.0171	9.0	2592	1.4043	0.7943
0.0171	10.0	2880	1.3356	0.8002
0.0154	11.0	3168	1.4528	0.7996
0.0154	12.0	3456	1.5514	0.7991
0.005	13.0	3744	1.6341	0.8046
0.0038	14.0	4032	1.6240	0.7984
0.0038	15.0	4320	1.7476	0.8014
0.0037	16.0	4608	1.6666	0.7982
0.0037	17.0	4896	1.7495	0.7950
0.0083	18.0	5184	1.6993	0.7932
0.0083	19.0	5472	1.6573	0.8077
0.002	20.0	5760	1.7430	0.7980
0.0012	21.0	6048	1.8135	0.7955
0.0012	22.0	6336	1.8316	0.7972
0.0022	23.0	6624	1.8717	0.7926
0.0022	24.0	6912	1.8183	0.7978
0.0014	25.0	7200	1.8361	0.7958

Framework versions

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