distilbert-base-uncased-finetuned-osdg

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: 1.8193
• F1 Score: 0.7962
• Accuracy: 0.8434

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: 5e-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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Score	Accuracy
0.3769	1.0	1017	0.8258	0.7729	0.8257
0.2759	2.0	2034	0.8364	0.7773	0.8262
0.1412	3.0	3051	1.0203	0.7833	0.8379
0.1423	4.0	4068	1.1603	0.7683	0.8224
0.0939	5.0	5085	1.3029	0.7843	0.8329
0.0757	6.0	6102	1.3562	0.7931	0.8379
0.0801	7.0	7119	1.2925	0.7840	0.8395
0.0311	8.0	8136	1.4632	0.7750	0.8318
0.0263	9.0	9153	1.5760	0.7843	0.8312
0.0196	10.0	10170	1.5689	0.7890	0.8417
0.0313	11.0	11187	1.6034	0.7909	0.8417
0.0007	12.0	12204	1.6725	0.7889	0.8406
0.0081	13.0	13221	1.6463	0.7911	0.8395
0.0061	14.0	14238	1.7730	0.7861	0.8345
0.003	15.0	15255	1.8001	0.7847	0.8379
0.0002	16.0	16272	1.7328	0.7912	0.8434
0.0	17.0	17289	1.7914	0.8011	0.8489
0.0009	18.0	18306	1.7772	0.7958	0.8456
0.0	19.0	19323	1.8028	0.7958	0.8434
0.0	20.0	20340	1.8193	0.7962	0.8434

Framework versions

• Transformers 4.19.2
• Pytorch 1.11.0+cu113
• Datasets 2.2.2
• Tokenizers 0.12.1