Skip to content

oleeviyababu/Image-Classification

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Image Classification

Training Machine learning model on the EuroSAT land cover classification dataset

CONSTRAINTS

  • Model's inability to learn complicated features or underfitting may be caused by a lack of layers and nodes.
  • The model's capacity to generalize to different datasets may be constrained by the fact that they were developed using only one dataset (EuroSAT).
  • The models are trained for a set number of epochs, which might not be enough to achieve their full potential.

IMPROVEMENTS

  • To boost the performance of the models, add more layers and nodes to make them deeper and more complicated.

  • To improve feature extraction and classification from larger datasets, apply transfer learning to pre-trained models.

  • Early stopping can be used to avoid overfitting and obtain peak performance without needing to train for a set number of epochs.

  • Use methods like cross-validation to enhance the models' accuracy and prevent bias brought on by the selection of the train-test split.

  • To combine many models for improved accuracy and resilience, use ensemble learning.

About

Training Machine learning model on the EuroSAT land cover classification dataset

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published