videomae-base-nicole

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: 0.8055
• Accuracy: 0.7969

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: 8
• eval_batch_size: 8
• 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: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.368	0.102	51	1.4286	0.2903
1.1282	1.102	102	1.0812	0.4516
0.8288	2.102	153	1.4727	0.4194
0.3846	3.102	204	1.1547	0.6774
0.3053	4.102	255	1.1199	0.6774
0.0898	5.102	306	0.6689	0.7742
0.304	6.102	357	0.6001	0.8065
0.1134	7.102	408	0.5982	0.8710
0.0641	8.102	459	0.4034	0.9032
0.0834	9.082	500	0.3761	0.9032

Framework versions

• Transformers 4.46.2
• Pytorch 2.0.1+cu117
• Datasets 3.1.0
• Tokenizers 0.20.3