videomae-base-finetuned-kinetics-0408_final_5sec_org_ab7_val_as123

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: 0.3219
• Accuracy: 0.9225

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: 65800

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.537	0.0100	659	0.3567	0.8740
0.0105	1.0100	1318	0.2819	0.9139
0.0039	2.0100	1977	0.3886	0.8858
0.0023	3.0100	2636	0.7164	0.8169
0.0222	4.0100	3295	0.2866	0.9264
0.0076	5.0100	3954	0.4126	0.9006
1.008	6.0100	4613	0.3480	0.9006
0.2086	7.0100	5272	0.2673	0.9311
0.4466	8.0100	5931	0.3282	0.9264
0.0067	9.0100	6590	0.3301	0.9116
1.3353	10.0100	7249	0.5289	0.8654
0.0003	11.0100	7908	0.6538	0.8732
0.0003	12.0100	8567	0.3669	0.9241
0.0184	13.0100	9226	0.3609	0.9171
0.0002	14.0100	9885	0.5037	0.8975
0.0001	15.0100	10544	0.4052	0.9202
0.0077	16.0100	11203	0.5142	0.9014
0.0001	17.0100	11862	0.3801	0.9241
0.0092	18.0100	12521	0.5826	0.9030
0.0012	19.0100	13180	0.4868	0.9155
0.0043	20.0100	13839	0.4497	0.9186
0.0001	21.0100	14498	0.7081	0.8858
0.0	22.0100	15157	0.8394	0.8725

Framework versions

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