distilbert-base-uncased_fold_1_ternary

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: 2.0582
• F1: 0.7326

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	0.5524	0.6755
0.5648	2.0	580	0.5654	0.7124
0.5648	3.0	870	0.6547	0.6896
0.2712	4.0	1160	0.8916	0.7263
0.2712	5.0	1450	1.1187	0.7120
0.1147	6.0	1740	1.2778	0.7114
0.0476	7.0	2030	1.4441	0.7151
0.0476	8.0	2320	1.5535	0.7133
0.0187	9.0	2610	1.6439	0.7212
0.0187	10.0	2900	1.7261	0.7313
0.0138	11.0	3190	1.6806	0.7292
0.0138	12.0	3480	1.8425	0.7111
0.009	13.0	3770	1.9207	0.7213
0.0045	14.0	4060	1.8900	0.7202
0.0045	15.0	4350	1.9730	0.7216
0.0042	16.0	4640	2.0775	0.7041
0.0042	17.0	4930	2.0514	0.7106
0.0019	18.0	5220	2.0582	0.7326
0.0039	19.0	5510	2.1010	0.7081
0.0039	20.0	5800	2.0487	0.7273
0.0025	21.0	6090	2.0415	0.7254
0.0025	22.0	6380	2.0753	0.7157
0.0017	23.0	6670	2.0554	0.7246
0.0017	24.0	6960	2.0644	0.7290
0.0001	25.0	7250	2.0711	0.7310

Framework versions

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