videomae-base-finetuned-deception-dataset-v2

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

• Loss: 1.3303
• Accuracy: 0.6790

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: 14
• eval_batch_size: 14
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 28
• optimizer: Use OptimizerNames.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: 252

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.1031	1.0	22	1.8863	0.5185
0.125	2.0	44	0.9140	0.8765
0.0816	3.0	66	1.2803	0.7037
0.097	4.0	88	1.3666	0.6914
0.0607	5.0	110	2.0708	0.5432
0.0848	6.0	132	1.5366	0.6049
0.0727	7.0	154	1.4673	0.6420
0.0716	8.0	176	1.3291	0.6914
0.1157	9.0	198	1.7054	0.5802
0.0738	10.0	220	1.2064	0.7160
0.0604	11.0	242	1.2631	0.6790
0.0509	11.4651	252	1.3303	0.6790

Framework versions

• Transformers 4.51.3
• Pytorch 2.1.0+cu121
• Datasets 3.5.0
• Tokenizers 0.21.1