self_harm_detection

This model is a fine-tuned version of Falconsai/nsfw_image_detection on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.0386
• Accuracy: 0.9860

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
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0772	0.9984	156	0.1007	0.9580
0.0351	1.9968	312	0.0557	0.9760
0.0206	2.9952	468	0.0386	0.9860

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1