distilbert-hate_speech18

This model is a fine-tuned version of agvidit1/DistilledBert_HateSpeech_pretrain on the hate_speech18 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4684
• Accuracy: 0.8588

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: 5.050180626898551e-06
• train_batch_size: 32
• eval_batch_size: 16
• seed: 33
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4404	1.0	240	0.4670	0.8533
0.4356	2.0	480	0.4642	0.8675
0.4303	3.0	720	0.4649	0.8748
0.4282	4.0	960	0.4694	0.8592
0.4273	5.0	1200	0.4638	0.8729
0.4256	6.0	1440	0.4651	0.8679
0.425	7.0	1680	0.4682	0.8560
0.4227	8.0	1920	0.4684	0.8588

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.1
• Datasets 2.15.0
• Tokenizers 0.15.0