videomae-base-finetuned-kinetics-0324-abnormal_val02_pretest_addnon

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

• Loss: 0.1758
• Accuracy: 0.9393

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: 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: 0.05
• training_steps: 90500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0211	0.0100	906	0.6697	0.7820
0.0326	1.0100	1812	0.7906	0.7976
0.0015	2.0100	2718	0.2768	0.9052
0.0013	3.0100	3624	0.2897	0.8924
0.0012	4.0100	4530	0.5914	0.8330
0.0014	5.0100	5436	0.5133	0.8677
0.0069	6.0100	6342	0.3602	0.8691
0.965	7.0100	7248	0.4386	0.8882
0.0142	8.0100	8154	0.3733	0.9016
0.9934	9.0100	9060	0.3805	0.8967
0.0019	10.0100	9966	0.3936	0.8995
0.161	11.0100	10872	0.3843	0.9066
0.0001	12.0100	11778	0.4220	0.8995
0.1619	13.0100	12684	0.2733	0.9342
0.0014	14.0100	13590	0.4517	0.9052
0.9858	15.0100	14496	0.4608	0.9030
0.0	16.0100	15402	0.3745	0.9193
0.0	17.0100	16308	0.3601	0.9257
0.0001	18.0100	17214	0.3452	0.9292
0.0006	19.0100	18120	0.3417	0.9278
0.0033	20.0100	19026	0.3737	0.9236
0.0092	21.0100	19932	0.4637	0.9186
0.0058	22.0100	20838	0.3807	0.9321
0.0001	23.0100	21744	0.4098	0.9299
0.0	24.0100	22650	0.4991	0.9165
0.0	25.0100	23556	0.3916	0.9278
0.2186	26.0100	24462	0.4615	0.9236
0.0015	27.0100	25368	0.5143	0.9179
0.0001	28.0100	26274	0.4412	0.9271

Framework versions

• Transformers 4.48.0
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0