distilbert-base-uncased_fold_11_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: 1.8389
• F1: 0.8057

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	0.4534	0.8011
0.4027	2.0	576	0.4299	0.8121
0.4027	3.0	864	0.4840	0.8142
0.1947	4.0	1152	0.7501	0.7992
0.1947	5.0	1440	1.0307	0.7866
0.0771	6.0	1728	1.1292	0.8034
0.0253	7.0	2016	1.2620	0.8033
0.0253	8.0	2304	1.4065	0.7954
0.0137	9.0	2592	1.4922	0.7887
0.0137	10.0	2880	1.4922	0.8050
0.0046	11.0	3168	1.4883	0.8097
0.0046	12.0	3456	1.5542	0.8133
0.0066	13.0	3744	1.5180	0.8000
0.0094	14.0	4032	1.6762	0.7919
0.0094	15.0	4320	1.5808	0.8005
0.0047	16.0	4608	1.7025	0.8012
0.0047	17.0	4896	1.6494	0.7986
0.0039	18.0	5184	1.7218	0.8010
0.0039	19.0	5472	1.8293	0.7994
0.0005	20.0	5760	1.8142	0.7980
0.0033	21.0	6048	1.8350	0.8037
0.0033	22.0	6336	1.8361	0.8042
0.0023	23.0	6624	1.8715	0.7996
0.0023	24.0	6912	1.8411	0.8057
0.0019	25.0	7200	1.8389	0.8057

Framework versions

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