Ir para o conteúdo
ModelListadosExterno
Baixar

VideoMAE Kinetics Wlasl100 CoR 30ep

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: 0.9436 Accuracy: 0.7692

Fontehuggingface.co/Shawon16/videoMAE_kinetics_wlasl100_coR_30ep/tree/8e4d635b09cffd39603e5c9cbc1fc1d464aa556f
TipoOtherModelo baseMCG-NJU/videomae-base-finetuned-kineticsVersãoRevision 8e4d635Cnota

Fonte do modelo

Tipo
Other
Tamanho do arquivo
329 MB
Modelo base
MCG-NJU/videomae-base-finetuned-kinetics
Versão
vRevision 8e4d635
Atualizado
24 de out. de 2026
Downloads
7
Fonte
Hugging Face

Descrição da fonte

videoMAE_kinetics_wlasl100_coR_30ep

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: 0.9436
  • 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: 5400
  • mixed_precision_training: Native AMP

Training results

Training LossEpochStepValidation LossAccuracy
18.48120.03331804.57350.0325
17.44881.03333603.98650.1272
13.73912.03325403.14020.4024
9.49913.03347212.36360.5444
6.2924.03339011.91270.6213
3.9935.033310811.56340.6923
2.4526.033212611.41930.6893
1.48817.033414421.28400.7101
0.90368.033316221.16050.7426
0.52819.033318021.12990.7101
0.322810.033219821.12830.7189
0.185411.033421630.95300.7426
0.130412.033323431.03240.7604
0.079713.033325230.99850.7456
0.0714.033227030.99050.7604
0.068115.033428840.96330.7574
0.055916.033330640.91130.7692
0.054817.033332440.92720.7544
0.036718.033234240.95050.7692
0.063619.033436050.92780.7633
0.076720.033337850.93330.7604
0.07121.033339650.94360.7692

Framework versions

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

Fontes

1 fonte
Hugging FacePrimárioVersão 8e4d635b09cffd39603e5c9cbc1fc1d464aa556f
huggingface.co/Shawon16/videoMAE_kinetics_wlasl100_coR_30ep/tree/8e4d635b09cffd39603e5c9cbc1fc1d464aa556f

Verificado 24 de out.

Artefatos de modelo

1 artefato

model.safetensors

safetensors · 329 MB · SHA-256 4c84a1678c84…3718 · Hugging Face

Download file

Trechos de fonte

3 trechos
Readme6 de ago.huggingface.co/Shawon16/videoMAE_kinetics_wlasl100_coR_30ep

--- library_name: transformers license: cc-by-nc-4.0 base_model: MCG-NJU/videomae-base-finetuned-kinetics tags: - generated_from_trainer metrics: - accuracy model-index: - name: videoMAE_kinetics_wlasl100_coR_30ep results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # videoMAE_kinetics_wlasl100_coR_30ep This model is a fine-tuned version of [MCG-NJU/videomae-base-finetuned-kinetics](https://huggingface.co/MCG-NJU/videomae-base-finetuned-kinetics) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.9436 - 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: 5400 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-------:|:----:|:---------------:|:--------:| | 18.4812 | 0.0333 | 180 | 4.5735 | 0.0325 | | 17.4488 | 1.0333 | 360 | 3.9865 | 0.1272 | | 13.7391 | 2.0332 | 540 | 3.1402 | 0.4024 | | 9.4991 | 3.0334 | 721 | 2.3636 | 0.5444 | | 6.292 | 4.0333 | 901 | 1.9127 | 0.6213 | | 3.993 | 5.0333 | 1081 | 1.5634 | 0.6923 | | 2.452 | 6.0332 | 1261 | 1.4193 | 0.6893 | | 1.4881 | 7.0334 | 1442 | 1.2840 | 0.7101 | | 0.9036 | 8.0333 | 1622 | 1.1605 | 0.7426 | | 0.5281 | 9.0333 | 1802 | 1.1299 | 0.7101 | | 0.3228 | 10.0332 | 1982 | 1.1283 | 0.7189 | | 0.1854 | 11.0334 | 2163 | 0.9530 | 0.7426 | | 0.1304 | 12.0333 | 2343 | 1.0324 | 0.7604 | | 0.0797 | 13.0333 | 2523 | 0.9985 | 0.7456 | | 0.07 | 14.0332 | 2703 | 0.9905 | 0.7604 | | 0.0681 | 15.0334 | 2884 | 0.9633 | 0.7574 | | 0.0559 | 16.0333 | 3064 | 0.9113 | 0.7692 | | 0.0548 | 17.0333 | 3244 | 0.9272 | 0.7544 | | 0.0367 | 18.0332 | 3424 | 0.9505 | 0.7692 | | 0.0636 | 19.0...

Note24 de out.huggingface.co/Shawon16/videoMAE_kinetics_wlasl100_coR_30ep

Source context: 7 downloads · 0 likes · Pipeline video-classification · Library transformers · Repo Shawon16/videoMAE_kinetics_wlasl100_coR_30ep

Note6 de ago.huggingface.co/Shawon16/videoMAE_kinetics_wlasl100_coR_30ep

Source context: 2 downloads · 0 likes · Pipeline video-classification · Library transformers · Repo Shawon16/videoMAE_kinetics_wlasl100_coR_30ep