distilbert-base-uncased_fold_5_binary

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: 0.5093
• F1: 0.7801

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	288	0.4760	0.7315
0.3992	2.0	576	0.4428	0.7785
0.3992	3.0	864	0.5093	0.7801
0.2021	4.0	1152	0.6588	0.7634
0.2021	5.0	1440	0.9174	0.7713
0.0945	6.0	1728	0.9832	0.7726
0.0321	7.0	2016	1.2103	0.7672
0.0321	8.0	2304	1.3759	0.7616
0.0134	9.0	2592	1.4405	0.7570
0.0134	10.0	2880	1.4591	0.7710
0.0117	11.0	3168	1.4947	0.7713
0.0117	12.0	3456	1.6224	0.7419
0.0081	13.0	3744	1.6462	0.7520
0.0083	14.0	4032	1.6880	0.7637
0.0083	15.0	4320	1.7080	0.7380
0.0048	16.0	4608	1.7352	0.7551
0.0048	17.0	4896	1.6761	0.7713
0.0024	18.0	5184	1.7553	0.76
0.0024	19.0	5472	1.7312	0.7673
0.005	20.0	5760	1.7334	0.7713
0.0032	21.0	6048	1.7963	0.7578
0.0032	22.0	6336	1.7529	0.7679
0.0025	23.0	6624	1.7741	0.7662
0.0025	24.0	6912	1.7515	0.7679
0.0004	25.0	7200	1.7370	0.7765

Framework versions

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