videomae-base-finetuned-sample_kine

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: 0.5079
• Accuracy: 0.8205

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: 140

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7564	0.1071	15	0.6660	0.6923
0.6614	1.1071	30	0.5677	0.6923
0.5941	2.1071	45	0.5079	0.8205
0.3661	3.1071	60	0.6175	0.6923
0.3258	4.1071	75	1.1649	0.7436
0.5887	5.1071	90	0.4697	0.7179
0.3907	6.1071	105	0.9874	0.6154
0.1948	7.1071	120	0.9959	0.6667
0.1424	8.1071	135	1.1357	0.6667
0.2198	9.0357	140	1.1467	0.6667

Framework versions

• Transformers 4.43.4
• Pytorch 2.4.1
• Datasets 2.20.0
• Tokenizers 0.19.1