VideoMAE_BdSLW60_FrameRate_NOT_Corrected_with_Augment_20_epoch_RQ

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: 0.0470
  • Accuracy: 0.9906
  • Precision: 0.9913
  • Recall: 0.9906
  • F1: 0.9906

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: 17940
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision Recall F1
11.7722 0.05 897 2.3329 0.4482 0.4616 0.4482 0.3915
2.7045 1.0500 1795 0.6715 0.8471 0.8870 0.8471 0.8384
0.7855 2.0500 2693 0.2378 0.9412 0.9474 0.9412 0.9401
0.5503 3.0500 3591 0.1367 0.9635 0.9686 0.9635 0.9635
0.2537 4.05 4488 0.1621 0.9612 0.9658 0.9612 0.9608
0.2549 5.0500 5386 0.1229 0.9765 0.9789 0.9765 0.9761
0.3236 6.0500 6284 0.0916 0.9765 0.9799 0.9765 0.9763
0.2078 7.0500 7182 0.1703 0.96 0.9647 0.96 0.9600
0.1967 8.05 8079 0.1708 0.9706 0.9731 0.9706 0.9707
0.2457 9.0500 8977 0.1500 0.9718 0.9772 0.9718 0.9716
0.0204 10.0500 9875 0.1181 0.9812 0.9833 0.9812 0.9811
0.0753 11.0500 10773 0.1418 0.9753 0.9775 0.9753 0.9755
0.0568 12.05 11670 0.1563 0.9765 0.9791 0.9765 0.9763
0.0851 13.0500 12568 0.0903 0.9847 0.9856 0.9847 0.9846
0.0106 14.0500 13466 0.0935 0.9871 0.9881 0.9871 0.9869
0.0171 15.0500 14364 0.0429 0.9929 0.9934 0.9929 0.9929
0.0025 16.05 15261 0.0584 0.9882 0.9890 0.9882 0.9882
0.0006 17.0500 16159 0.0693 0.9882 0.9894 0.9882 0.9883
0.0001 18.0500 17057 0.0513 0.9906 0.9913 0.9906 0.9906
0.0001 19.0492 17940 0.0470 0.9906 0.9913 0.9906 0.9906

Framework versions

  • Transformers 4.46.1
  • Pytorch 2.5.1+cu124
  • Datasets 3.1.0
  • Tokenizers 0.20.1
Downloads last month
11
Safetensors
Model size
86.3M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and the model is not deployed on the HF Inference API.

Model tree for Shawon16/VideoMAE_BdSLW60_FrameRate_NOT_Corrected_with_Augment_20_epoch_RQ

Finetuned
(480)
this model