distilbert-base-uncased_fold_5_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: 1.8096
• F1: 0.7352

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	0.6322	0.6742
0.5533	2.0	582	0.5861	0.7285
0.5533	3.0	873	0.6893	0.7117
0.2576	4.0	1164	1.0393	0.7124
0.2576	5.0	1455	1.1506	0.6988
0.1097	6.0	1746	1.3005	0.7166
0.0487	7.0	2037	1.5242	0.7124
0.0487	8.0	2328	1.5705	0.7010
0.0253	9.0	2619	1.5180	0.7194
0.0253	10.0	2910	1.6251	0.7062
0.022	11.0	3201	1.6299	0.7169
0.022	12.0	3492	1.7322	0.7091
0.0065	13.0	3783	1.8441	0.7044
0.0093	14.0	4074	1.9063	0.7097
0.0093	15.0	4365	1.8096	0.7352
0.0037	16.0	4656	1.8589	0.7321
0.0037	17.0	4947	1.9687	0.7211
0.0036	18.0	5238	1.9244	0.7285
0.0045	19.0	5529	1.9835	0.7299
0.0045	20.0	5820	2.0766	0.7139
0.0024	21.0	6111	2.1118	0.7144
0.0024	22.0	6402	2.0544	0.7197
0.0006	23.0	6693	2.0914	0.7217
0.0006	24.0	6984	2.1028	0.7195
0.0006	25.0	7275	2.1174	0.7224

Framework versions

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