distilbert-base-uncased_fold_13_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: 1.7433
• F1: 0.8138

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	0.4101	0.8087
0.4128	2.0	582	0.4605	0.8197
0.4128	3.0	873	0.5011	0.8130
0.1997	4.0	1164	0.6882	0.8147
0.1997	5.0	1455	0.9653	0.8092
0.0913	6.0	1746	1.1020	0.8031
0.0347	7.0	2037	1.2687	0.8050
0.0347	8.0	2328	1.2383	0.8103
0.0173	9.0	2619	1.3631	0.8066
0.0173	10.0	2910	1.4282	0.8001
0.0104	11.0	3201	1.4410	0.8179
0.0104	12.0	3492	1.5318	0.8018
0.0063	13.0	3783	1.5866	0.8018
0.0043	14.0	4074	1.4987	0.8159
0.0043	15.0	4365	1.6275	0.8181
0.0048	16.0	4656	1.5811	0.8231
0.0048	17.0	4947	1.6228	0.8182
0.0048	18.0	5238	1.7235	0.8138
0.0055	19.0	5529	1.7018	0.8066
0.0055	20.0	5820	1.7340	0.8069
0.0046	21.0	6111	1.7143	0.8156
0.0046	22.0	6402	1.7367	0.8159
0.0037	23.0	6693	1.7551	0.8151
0.0037	24.0	6984	1.7479	0.8145
0.0009	25.0	7275	1.7433	0.8138

Framework versions

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