videomae-base-finetuned-xd-violence-binary

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

• Loss: 0.6208
• Accuracy: 0.6674
• F1: 0.7483
• Precision: 0.7108
• Recall: 0.7900
• Specificity: 0.4625
• True Positives: 17801
• True Negatives: 6233
• False Positives: 7244
• False Negatives: 4733

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: 0.0005
• train_batch_size: 8
• eval_batch_size: 8
• 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
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 1684

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall	Specificity	True Positives	True Negatives	False Positives	False Negatives
0.6787	0.2506	422	0.7039	0.5679	0.5938	0.7211	0.5047	0.6736	11374	9078	4399	11160
0.6389	1.2506	844	0.6784	0.5828	0.5991	0.7513	0.4981	0.7243	11225	9761	3716	11309
0.6126	2.2506	1266	0.6322	0.6599	0.7373	0.7135	0.7626	0.4881	17185	6578	6899	5349
0.6762	3.2482	1684	0.6208	0.6674	0.7483	0.7108	0.7900	0.4625	17801	6233	7244	4733

Framework versions

• Transformers 4.51.3
• Pytorch 2.1.0+cu118
• Datasets 3.6.0
• Tokenizers 0.21.1