videomae-base-finetuned-kinetics-final-contest

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: 0.6795
• Accuracy: 0.8458

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: 9e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 2464

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0054	0.0913	225	0.9635	0.7764
0.3981	1.0913	450	0.7286	0.7747
0.1595	2.0913	675	0.6008	0.8215
0.0302	3.0913	900	0.7152	0.8024
0.015	4.0913	1125	0.7613	0.7903
0.0019	5.0913	1350	0.6450	0.8354
0.0037	6.0913	1575	0.5948	0.8510
0.0037	7.0913	1800	0.6545	0.8354
0.0005	8.0913	2025	0.6526	0.8440
0.0005	9.0913	2250	0.6803	0.8458
0.0004	10.0869	2464	0.6795	0.8458

Framework versions

• Transformers 4.40.1
• Pytorch 2.3.0+cu121
• Datasets 2.18.0
• Tokenizers 0.19.1