ALL_NoCrop_Aug16F-8B16F-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: 0.5898
• Accuracy: 0.8233

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• 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: 2308

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6192	0.0624	144	0.6215	0.6829
0.3725	1.0624	288	0.5207	0.7530
0.2718	2.0624	432	0.6017	0.7744
0.1424	3.0624	576	0.5600	0.7774
0.1755	4.0624	720	0.6294	0.7896
0.1333	5.0624	864	0.6707	0.7957
0.0937	6.0624	1008	0.8023	0.7927
0.0692	7.0624	1152	0.7964	0.8079
0.0994	8.0624	1296	0.8140	0.7988
0.028	9.0624	1440	0.8548	0.8049
0.0692	10.0624	1584	0.9202	0.8018
0.027	11.0624	1728	0.9143	0.7927

Framework versions

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