Ir para o conteúdo
ModelListadosExterno
Baixar

Timesformer Wlasl 100 200ep CoR

This model is a fine-tuned version of facebook/timesformer-base-finetuned-k400 on an unknown dataset. It achieves the following results on the evaluation set: Loss: 2.5784 Accuracy: 0.5592

Fontehuggingface.co/Shawon16/timesformer_wlasl_100_200ep_coR_/tree/7eed1a04390044b6254920c5966e8ee0a6fce4da
TipoOtherModelo basefacebook/timesformer-base-finetuned-k400VersãoModelBnota

Fonte do modelo

Tipo
Other
Tamanho do arquivo
463 MB
Modelo base
facebook/timesformer-base-finetuned-k400
Versão
vModel
Atualizado
24 de out. de 2026
Fonte
Hugging Face

Descrição da fonte

timesformer_wlasl_100_200ep_coR_

This model is a fine-tuned version of facebook/timesformer-base-finetuned-k400 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.5784
  • Accuracy: 0.5592

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: 36000
  • mixed_precision_training: Native AMP

Training results

Training LossEpochStepValidation LossAccuracy
18.90030.0051804.68680.0089
18.36931.00503604.53530.0414
17.24762.00505404.28350.0828
15.30513.00507213.89920.1598
12.95924.0059013.48290.2456
10.49345.005010813.15370.3491
8.25886.005012612.81150.4172
6.17657.005014422.56710.4438
4.3218.00516222.27920.5030
2.8749.005018022.10450.5296
1.763410.005019821.97380.5414
1.089111.005021631.89310.5592
0.699212.00523431.76060.5651
0.42913.005025231.68790.5799
0.32714.005027031.70750.5976
0.195915.005028841.64860.5917
0.158816.00530641.84400.5621
0.156317.005032441.73940.5888
0.191318.005034241.90170.5710
0.160319.005036051.95270.5828
0.100720.00537851.90930.6124
0.095221.005039651.80470.6036
0.083322.005041451.96310.5680
0.153723.005043261.92070.6154
0.219824.00545061.89790.5976
0.131625.005046861.91790.5917
0.150326.005048661.97430.5769
0.142927.005050472.04550.5740
0.098828.00552272.08440.5710
0.181929.005054072.05160.5947
0.118430.005055872.40840.5621
0.111331.005057682.27450.5473
0.138232.00559482.55050.5385
0.137533.005061282.29800.5592
0.241434.005063082.30810.5325
0.125535.005064892.13510.5858
0.102436.00566692.19040.6006
0.163937.005068492.25390.5740
0.162538.005070292.13300.5858
0.221639.005072102.41280.5503
0.164840.00573902.42920.5680
0.127641.005075702.37370.6036
0.087242.005077502.22010.6183
0.136643.005079312.35840.5917
0.069744.00581112.25360.5858
0.10945.005082912.13990.6095
0.111446.005084712.28590.5740
0.104347.005086522.42370.5592
0.099948.00588322.48050.5385
0.218449.005090122.30020.5740
0.177650.005091922.34140.5533
0.122251.005093732.38390.5888
0.068752.00595532.59900.5592
0.08653.005097332.60020.5799
0.040354.005099132.31520.5947
0.060355.0050100942.44020.5858
0.04456.005102742.34510.6124
0.119357.0050104542.51410.5680
0.106858.0050106342.64380.5828
0.072559.0050108152.51380.5858
0.11960.005109952.51380.5888
0.192761.0050111752.77760.5503
0.145862.0050113552.57840.5592

Framework versions

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

Fontes

1 fonte
Hugging FacePrimárioVersão 7eed1a04390044b6254920c5966e8ee0a6fce4da
huggingface.co/Shawon16/timesformer_wlasl_100_200ep_coR_/tree/7eed1a04390044b6254920c5966e8ee0a6fce4da

Verificado 24 de out.

Artefatos de modelo

1 artefato

model.safetensors

safetensors · 463 MB · SHA-256 f416aae3b902…3657 · Hugging Face

Download file

Trechos de fonte

2 trechos
Readme6 de ago.huggingface.co/Shawon16/timesformer_wlasl_100_200ep_coR_

--- library_name: transformers license: cc-by-nc-4.0 base_model: facebook/timesformer-base-finetuned-k400 tags: - generated_from_trainer metrics: - accuracy model-index: - name: timesformer_wlasl_100_200ep_coR_ 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. --> # timesformer_wlasl_100_200ep_coR_ This model is a fine-tuned version of [facebook/timesformer-base-finetuned-k400](https://huggingface.co/facebook/timesformer-base-finetuned-k400) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 2.5784 - Accuracy: 0.5592 ## 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: 36000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-------:|:-----:|:---------------:|:--------:| | 18.9003 | 0.005 | 180 | 4.6868 | 0.0089 | | 18.3693 | 1.0050 | 360 | 4.5353 | 0.0414 | | 17.2476 | 2.0050 | 540 | 4.2835 | 0.0828 | | 15.3051 | 3.0050 | 721 | 3.8992 | 0.1598 | | 12.9592 | 4.005 | 901 | 3.4829 | 0.2456 | | 10.4934 | 5.0050 | 1081 | 3.1537 | 0.3491 | | 8.2588 | 6.0050 | 1261 | 2.8115 | 0.4172 | | 6.1765 | 7.0050 | 1442 | 2.5671 | 0.4438 | | 4.321 | 8.005 | 1622 | 2.2792 | 0.5030 | | 2.874 | 9.0050 | 1802 | 2.1045 | 0.5296 | | 1.7634 | 10.0050 | 1982 | 1.9738 | 0.5414 | | 1.0891 | 11.0050 | 2163 | 1.8931 | 0.5592 | | 0.6992 | 12.005 | 2343 | 1.7606 | 0.5651 | | 0.429 | 13.0050 | 2523 | 1.6879 | 0.5799 | | 0.327 | 14.0050 | 2703 | 1.7075 | 0.5976 | | 0.1959 | 15.0050 | 2884 | 1.6486 | 0.5917 | | 0.1588 | 16.005 | 3064 | 1.8440 | 0.5621 | | 0.1563 | 17.0050 | 3244 | 1.7394 | 0.5888 | | 0.1913 | 18.0050 | 3424 | 1.9017 | 0.5710 | | 0.1603 | 19.0050 |...

Note6 de ago.huggingface.co/Shawon16/timesformer_wlasl_100_200ep_coR_

Source context: 8 downloads · 0 likes · Pipeline video-classification · Library transformers · Repo Shawon16/timesformer_wlasl_100_200ep_coR_