RALL_RGBCROP_Aug16F-8B16F-lr5e6_2

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: 0.6144
• Accuracy: 0.8554

1.Best Learning Rate lr : 5e-6

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-06
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• 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: 3462

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4481	0.0835	289	0.5643	0.7485
0.203	1.0835	578	0.5270	0.8016
0.0444	2.0835	867	0.6951	0.8139
0.0038	3.0835	1156	0.8074	0.8282
0.0041	4.0835	1445	0.8784	0.8344
0.0005	5.0835	1734	0.9289	0.8344
0.0004	6.0835	2023	0.9893	0.8241
0.0003	7.0835	2312	1.0132	0.8323
0.0003	8.0835	2601	1.0356	0.8282

Framework versions

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