videomae-base-finetuned-kinetics-0403_final_only5sec

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: 0.9866
• Accuracy: 0.7954

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: 62100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0258	0.0100	622	0.3408	0.8950
0.0148	1.0100	1244	0.3555	0.8603
0.0013	2.0100	1866	0.5279	0.8754
0.0044	3.0100	2488	0.8037	0.8283
0.0127	4.0100	3110	0.6375	0.8461
1.0048	5.0100	3732	1.2457	0.7811
0.0006	6.0100	4354	0.7576	0.8292
0.0003	7.0100	4976	0.4766	0.8968
0.0027	8.0100	5598	0.6147	0.8496
0.0012	9.0100	6220	0.6214	0.8345
0.0157	10.0100	6842	0.6998	0.8514
0.0099	11.0100	7464	1.1907	0.7847
0.0537	12.0100	8086	1.2955	0.8043
0.0001	13.0100	8708	1.1235	0.7972
0.0007	14.0100	9330	1.3138	0.7900
0.0028	15.0100	9952	1.0310	0.8221
0.0008	16.0100	10574	0.8199	0.8514
0.015	17.0100	11196	0.5774	0.8852
0.0002	18.0100	11818	0.9953	0.8407
0.6463	19.0100	12440	1.2704	0.8132
0.0001	20.0100	13062	1.0214	0.8203
0.0012	21.0100	13684	1.2372	0.8194
0.0001	22.0100	14306	1.0967	0.8372

Framework versions

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