Skip to content

Latest commit

 

History

History
8 lines (5 loc) · 543 Bytes

README.md

File metadata and controls

8 lines (5 loc) · 543 Bytes

Intel-Image-Classification

Multiclass classification of image scenes on data posted by Intel for image classification challenge

  • The dataset contains real world images of different scenes like mountains,forests,glaciers,streets,buildings etc. (data source : 'https://www.kaggle.com/puneet6060/intel-image-classification').

  • Moulded Moblienet V2 (trained on imagenentdataset) and retrained on the Intel-Image-Classification dataset.

  • Preprocessed and trained using tensorflow and keras and achieved accuracy of 89% validation accuracy.