distilbert-base-uncased_fold_8_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: 1.6283
• F1: 0.8178

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	290	0.4038	0.7981
0.409	2.0	580	0.4023	0.8176
0.409	3.0	870	0.5245	0.8169
0.1938	4.0	1160	0.6242	0.8298
0.1938	5.0	1450	0.8432	0.8159
0.0848	6.0	1740	1.0887	0.8015
0.038	7.0	2030	1.0700	0.8167
0.038	8.0	2320	1.0970	0.8241
0.0159	9.0	2610	1.2474	0.8142
0.0159	10.0	2900	1.3453	0.8184
0.01	11.0	3190	1.4412	0.8147
0.01	12.0	3480	1.4263	0.8181
0.007	13.0	3770	1.3859	0.8258
0.0092	14.0	4060	1.4633	0.8128
0.0092	15.0	4350	1.4304	0.8206
0.0096	16.0	4640	1.5081	0.8149
0.0096	17.0	4930	1.5239	0.8189
0.0047	18.0	5220	1.5268	0.8151
0.0053	19.0	5510	1.5445	0.8173
0.0053	20.0	5800	1.6051	0.8180
0.0014	21.0	6090	1.5981	0.8211
0.0014	22.0	6380	1.5957	0.8225
0.001	23.0	6670	1.5838	0.8189
0.001	24.0	6960	1.6301	0.8178
0.0018	25.0	7250	1.6283	0.8178

Framework versions

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