brand-safety-model

This model is a fine-tuned version of distilbert/distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5795
• Accuracy: 0.8630

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: 3e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.435	1.0	106	1.1905	0.7450
0.8006	2.0	212	0.7220	0.8152
0.5227	3.0	318	0.5708	0.8406
0.4252	4.0	424	0.5011	0.8501
0.3078	5.0	530	0.4905	0.8506
0.2471	6.0	636	0.5174	0.8447
0.1752	7.0	742	0.5095	0.8589
0.2241	8.0	848	0.5265	0.8524
0.1165	9.0	954	0.5525	0.8577
0.102	10.0	1060	0.5510	0.8512
0.0691	11.0	1166	0.5730	0.8577
0.0809	12.0	1272	0.5670	0.8619
0.0631	13.0	1378	0.5715	0.8625
0.0589	14.0	1484	0.5756	0.8625
0.0628	15.0	1590	0.5795	0.8630

Framework versions

• Transformers 4.46.3
• Pytorch 2.4.0
• Datasets 2.20.0
• Tokenizers 0.20.3