RALL_MVCROP-ori16Ftest_5e6-poly

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: 1.0783
• Accuracy: 0.6220

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: polynomial
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 1152

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6828	0.0833	96	0.6977	0.5244
0.692	1.0833	192	0.7108	0.5244
0.5597	2.0833	288	0.7554	0.5732
0.5573	3.0833	384	0.6685	0.5610
0.5035	4.0833	480	0.8520	0.5549
0.4499	5.0833	576	0.8314	0.5976
0.3592	6.0833	672	0.8039	0.5976
0.3792	7.0833	768	0.7147	0.6341
0.2453	8.0833	864	0.7976	0.6341
0.1699	9.0833	960	1.1120	0.5915
0.19	10.0833	1056	1.0033	0.6159
0.1862	11.0833	1152	1.0783	0.6220

Framework versions

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