distilbert-base-uncased_fold_6_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: 1.7209
• F1: 0.8156

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	0.4115	0.8048
0.3976	2.0	580	0.3980	0.8156
0.3976	3.0	870	0.5953	0.8142
0.1965	4.0	1160	0.7940	0.8057
0.1965	5.0	1450	0.8098	0.8069
0.0847	6.0	1740	1.0293	0.7913
0.03	7.0	2030	1.1649	0.8073
0.03	8.0	2320	1.2876	0.7973
0.0166	9.0	2610	1.3260	0.8038
0.0166	10.0	2900	1.3523	0.8084
0.0062	11.0	3190	1.3814	0.8097
0.0062	12.0	3480	1.4134	0.8165
0.0113	13.0	3770	1.5374	0.8068
0.006	14.0	4060	1.5808	0.8100
0.006	15.0	4350	1.6551	0.7972
0.0088	16.0	4640	1.5793	0.8116
0.0088	17.0	4930	1.6134	0.8143
0.0021	18.0	5220	1.6204	0.8119
0.0031	19.0	5510	1.7006	0.8029
0.0031	20.0	5800	1.6777	0.8145
0.0019	21.0	6090	1.7202	0.8079
0.0019	22.0	6380	1.7539	0.8053
0.0008	23.0	6670	1.7408	0.8119
0.0008	24.0	6960	1.7388	0.8176
0.0014	25.0	7250	1.7209	0.8156

Framework versions

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