CoolFace
Modelpublic

pavitemple/finetuned-Accident-MultipleLabels-Video-subset

sourceHugging Facecc-by-nc-4.0updated 3y agoView on Hugging Face
0likes7downloads
Model Card

<!-- 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. -->

finetuned-Accident-MultipleLabels-Video-subset

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: 2.0208
  • —Accuracy: 0.2593

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
  • —trainbatchsize: 4
  • —evalbatchsize: 4
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —lrschedulerwarmup_ratio: 0.1
  • —training_steps: 200

Training results

Training LossEpochStepValidation LossAccuracy
1.91930.08161.91080.2344
1.40251.08322.09750.1094
1.54132.08482.03030.2969
1.41473.08642.34470.2969
1.38044.08802.47330.0469
1.36875.08962.40510.3438
1.29296.081122.36910.3125
1.16617.081282.64120.1875
1.20248.081442.64270.3125
1.08249.081602.74780.3594
1.101110.081762.77890.3594
0.949211.081922.77060.3281
1.123212.042002.79440.2969

Framework versions

  • —Transformers 4.35.0
  • —Pytorch 2.1.0
  • —Datasets 2.14.6
  • —Tokenizers 0.14.1