videomae-base-finetuned-kinetics-allkisa-0219

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: 0.3020
• Accuracy: 0.9394

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: 7840

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0617	0.0251	197	0.3995	0.8433
0.0008	1.0251	394	0.3034	0.8715
0.0006	2.0251	591	0.3897	0.8762
1.2877	3.0251	788	0.3241	0.8809
0.0024	4.0251	985	0.3393	0.8856
0.0002	5.0251	1182	0.3336	0.9013
0.0005	6.0251	1379	0.4018	0.9028
0.0001	7.0251	1576	0.4943	0.8966
0.0004	8.0251	1773	0.6253	0.8323
0.0003	9.0251	1970	0.3540	0.9075
0.0005	10.0251	2167	0.4543	0.9013
0.0003	11.0251	2364	0.6340	0.8746
0.0	12.0251	2561	0.4580	0.9028
0.0004	13.0251	2758	0.5239	0.8997
0.0003	14.0251	2955	0.4695	0.9028
0.0025	15.0251	3152	0.6634	0.8762
0.0001	16.0251	3349	0.5013	0.9075
0.0	17.0251	3546	0.4318	0.9216
0.0001	18.0251	3743	0.4857	0.9169
0.0001	19.0251	3940	0.4436	0.9138
0.0002	20.0251	4137	0.5982	0.8950
0.0059	21.0251	4334	0.4260	0.9248
0.0376	22.0251	4531	0.4158	0.9263
0.0	23.0251	4728	0.4742	0.9169
0.0	24.0251	4925	0.4715	0.9107
0.0002	25.0251	5122	0.4703	0.9044
0.0004	26.0251	5319	0.4396	0.9138
0.0001	27.0251	5516	0.4424	0.9201
0.0	28.0251	5713	0.4931	0.8966
0.0001	29.0251	5910	0.5090	0.9028
0.0001	30.0251	6107	0.5055	0.9028
0.0	31.0251	6304	0.5457	0.8934
0.0227	32.0251	6501	0.5026	0.9013

Framework versions

• Transformers 4.48.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0