distilbert-base-uncased_fold_12_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: 1.7046
• F1: 0.8165

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	0.4165	0.7983
0.4052	2.0	580	0.4005	0.8213
0.4052	3.0	870	0.6003	0.8078
0.1906	4.0	1160	0.8181	0.7945
0.1906	5.0	1450	0.7775	0.7955
0.0853	6.0	1740	1.0667	0.7912
0.0407	7.0	2030	1.2061	0.7907
0.0407	8.0	2320	1.2522	0.8011
0.0145	9.0	2610	1.3073	0.8110
0.0145	10.0	2900	1.4895	0.7994
0.015	11.0	3190	1.4568	0.8082
0.015	12.0	3480	1.4883	0.8058
0.005	13.0	3770	1.4334	0.8217
0.0026	14.0	4060	1.5032	0.8255
0.0026	15.0	4350	1.5694	0.8193
0.0062	16.0	4640	1.6058	0.8105
0.0062	17.0	4930	1.7390	0.8058
0.0051	18.0	5220	1.6942	0.8100
0.0012	19.0	5510	1.6891	0.8151
0.0012	20.0	5800	1.6961	0.8132
0.0007	21.0	6090	1.6793	0.8168
0.0007	22.0	6380	1.7542	0.8077
0.0027	23.0	6670	1.6869	0.8203
0.0027	24.0	6960	1.7006	0.8194
0.0028	25.0	7250	1.7046	0.8165

Framework versions

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