distilbert-base-uncased_fold_5_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: 2.1368
• F1: 0.7682

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	0.6423	0.7465
0.5563	2.0	582	0.6001	0.7631
0.5563	3.0	873	0.6884	0.7785
0.2595	4.0	1164	0.9920	0.7439
0.2595	5.0	1455	1.1434	0.7631
0.1159	6.0	1746	1.3289	0.7606
0.0473	7.0	2037	1.3966	0.7708
0.0473	8.0	2328	1.4761	0.7606
0.0282	9.0	2619	1.6144	0.7542
0.0282	10.0	2910	1.5642	0.7695
0.0134	11.0	3201	1.7206	0.7593
0.0134	12.0	3492	1.8008	0.7542
0.0059	13.0	3783	1.8056	0.7746
0.002	14.0	4074	1.9160	0.7593
0.002	15.0	4365	2.0223	0.7606
0.0052	16.0	4656	1.9112	0.7810
0.0052	17.0	4947	1.9040	0.7772
0.0056	18.0	5238	1.9852	0.7734
0.0061	19.0	5529	2.0590	0.7644
0.0061	20.0	5820	2.1078	0.7631
0.0044	21.0	6111	2.1177	0.7631
0.0044	22.0	6402	2.0983	0.7644
0.0012	23.0	6693	2.1384	0.7670
0.0012	24.0	6984	2.1467	0.7657
0.0018	25.0	7275	2.1368	0.7682

Framework versions

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