videomae-base-finetuned-ucfsuc-subset

This model is a fine-tuned version of MCG-NJU/videomae-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6334
• Accuracy: 0.7079

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: 4
• eval_batch_size: 4
• 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: 0.1
• training_steps: 310

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7131	0.1032	32	0.7573	0.3596
0.7351	1.1032	64	0.6929	0.5393
0.6892	2.1032	96	0.7539	0.5393
0.693	3.1032	128	0.7079	0.5169
0.5903	4.1032	160	0.7080	0.6180
0.626	5.1032	192	0.6610	0.6854
0.6316	6.1032	224	0.5789	0.7303
0.5287	7.1032	256	0.6366	0.7079
0.6648	8.1032	288	0.6215	0.7191
0.4605	9.0710	310	0.6334	0.7079

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3