distilbert-base-uncased_fold_7_ternary_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: 2.0462
• F1: 0.7836

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	291	0.5719	0.7490
0.5541	2.0	582	0.5563	0.7836
0.5541	3.0	873	0.7301	0.7849
0.2509	4.0	1164	0.8073	0.7926
0.2509	5.0	1455	1.0842	0.7823
0.1182	6.0	1746	1.1721	0.7900
0.0537	7.0	2037	1.4060	0.7785
0.0537	8.0	2328	1.4497	0.7836
0.0262	9.0	2619	1.4722	0.7708
0.0262	10.0	2910	1.6529	0.7772
0.0131	11.0	3201	1.6573	0.7862
0.0131	12.0	3492	1.6986	0.7823
0.0115	13.0	3783	1.7765	0.7810
0.0098	14.0	4074	1.8036	0.7862
0.0098	15.0	4365	1.7684	0.7926
0.0028	16.0	4656	1.8385	0.7836
0.0028	17.0	4947	1.7903	0.7887
0.0054	18.0	5238	1.9065	0.7810
0.0007	19.0	5529	1.9331	0.7875
0.0007	20.0	5820	1.9384	0.7849
0.0006	21.0	6111	1.8687	0.7887
0.0006	22.0	6402	2.0603	0.7785
0.0009	23.0	6693	2.0403	0.7836
0.0009	24.0	6984	2.0348	0.7810
0.0005	25.0	7275	2.0462	0.7836

Framework versions

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