videomae-base-Badminton_strokes-finetuned-stroke-classification_10

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: 0.5231
• Accuracy: 0.8551

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: 24
• eval_batch_size: 24
• 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: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 2450

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.3775	0.1004	246	1.5477	0.4784
0.9098	1.1004	492	1.2118	0.6965
0.6294	2.1004	738	1.1314	0.7164
0.6993	3.1004	984	1.1107	0.7333
0.4355	4.1004	1230	1.0774	0.7497
0.3388	5.1004	1476	1.0708	0.7673
0.448	6.1004	1722	1.1309	0.7678
0.2328	7.1004	1968	1.0999	0.7737
0.2329	8.1004	2214	1.1141	0.7807
0.1596	9.0963	2450	1.1185	0.7795

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0