VideoMAE_kinetics_wlasl100_20epoch_Signers

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: 0.9211
• Accuracy: 0.7692

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• 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: 3600
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
18.4524	0.05	180	4.5268	0.0325
16.7504	1.0499	360	3.6888	0.1864
12.2073	2.0499	540	2.8001	0.4675
8.0525	3.0501	721	2.1476	0.6006
5.2821	4.05	901	1.7964	0.6450
3.2722	5.0499	1081	1.4956	0.7041
2.0177	6.0499	1261	1.3690	0.6746
1.2351	7.0501	1442	1.2708	0.6923
0.7437	8.05	1622	1.1608	0.7160
0.4375	9.0499	1802	1.0340	0.7515
0.2713	10.0499	1982	1.1279	0.7041
0.1569	11.0501	2163	0.9909	0.7367
0.1114	12.05	2343	0.9958	0.7456
0.074	13.0499	2523	0.9627	0.7485
0.0588	14.0499	2703	0.9344	0.7604
0.0579	15.0501	2884	0.9531	0.7663
0.0511	16.05	3064	0.9183	0.7663
0.0456	17.0499	3244	0.9359	0.7663
0.0399	18.0499	3424	0.9284	0.7633
0.0429	19.0487	3600	0.9211	0.7692

Framework versions

• Transformers 4.46.1
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.1