Skip to content

dsp-uga/hestu-p2

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

hestu-p2

Models

Our team developed two models to solve this classification problem

  • Decision Tree Method

  • Transfer learning ResNet Model

ResNet Model

For this model our team has yet to be able to train on the full training test. As of now, our team does not have any GCP credits left, However, our team was able to locally download over 400,000 for the 800,000 and create a preliminary training and testing set of our own from concatenating and splitting the provided CSV’s. Using this data we obtained an accuracy of 85%

How to run the ResNet model: open the Jupyter Notebook titled “resNet” and add and run the following line:

Tain_ResNet(path2trainCSV, val_fract, path2ims, num_epochs=25)
  • path2trainCSV: a path to the location of the training csv

  • val_fract: percent of training set used for validation

  • path2ims: a path to the location of the image data

Contributions

Please see our CONTRIBUTORS file for more details.

Authors

License

This project is licensed under the MIT License - see the LICENSE file for the details.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published