distilbert-base-uncased_fold_2_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: 1.8941
• F1: 0.7889

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	294	0.6025	0.7402
0.5688	2.0	588	0.5025	0.7943
0.5688	3.0	882	0.6102	0.7794
0.2582	4.0	1176	0.8896	0.7835
0.2582	5.0	1470	1.0392	0.7821
0.1185	6.0	1764	1.0865	0.7848
0.0461	7.0	2058	1.2951	0.7686
0.0461	8.0	2352	1.3348	0.7821
0.0313	9.0	2646	1.4267	0.7876
0.0313	10.0	2940	1.4004	0.7957
0.0142	11.0	3234	1.5501	0.7794
0.0083	12.0	3528	1.5564	0.7903
0.0083	13.0	3822	1.5699	0.7876
0.0067	14.0	4116	1.7725	0.7794
0.0067	15.0	4410	1.7642	0.7767
0.0031	16.0	4704	1.7891	0.7848
0.0031	17.0	4998	1.8528	0.7740
0.0054	18.0	5292	1.8378	0.7781
0.003	19.0	5586	1.8223	0.7862
0.003	20.0	5880	1.7935	0.7930
0.0021	21.0	6174	1.9117	0.7808
0.0021	22.0	6468	1.8891	0.7930
0.0015	23.0	6762	1.9167	0.7916
0.0006	24.0	7056	1.9193	0.7862
0.0006	25.0	7350	1.8941	0.7889

Framework versions

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