ALL_RGBCROP_Aug16F-16B16F-GACWDlr

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: 0.6816
• Accuracy: 0.8463

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.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: 0.1
• training_steps: 1728

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6361	0.0417	72	0.6432	0.6382
0.3554	1.0417	144	0.5090	0.7439
0.1632	2.0417	216	0.5323	0.7825
0.0439	3.0417	288	0.6035	0.8069
0.0044	4.0417	360	0.8301	0.7927
0.0028	5.0417	432	0.8714	0.8110
0.0008	6.0417	504	0.9483	0.8089
0.0005	7.0417	576	0.9650	0.8191
0.0005	8.0417	648	0.9847	0.8089
0.0005	9.0417	720	1.0961	0.8008
0.0003	10.0417	792	1.0523	0.8110
0.0003	11.0417	864	1.0718	0.8171
0.0002	12.0417	936	1.0848	0.8130

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1