videomae-base-finetuned-TikHaram

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: 0.7397
• Accuracy: 0.6887

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 295

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1138	0.1017	30	0.9141	0.7169
0.6999	1.1017	60	0.6574	0.7877
0.4804	2.1017	90	0.5710	0.7961
0.2795	3.1017	120	0.7722	0.7169
0.1862	4.1017	150	0.6860	0.7877
0.162	5.1017	180	0.6509	0.7810
0.1456	6.1017	210	0.7687	0.7574
0.1351	7.1017	240	0.7653	0.7675
0.0798	8.1017	270	0.7602	0.7751
0.246	9.0847	295	0.7605	0.7734

Framework versions

• Transformers 4.42.3
• Pytorch 2.1.2
• Datasets 2.20.0
• Tokenizers 0.19.1