Analysis on Transfer Learning
- 0 Collaborators
Conducted analysis on the accuracy achieved using Transfer Learning by retraining the last layer of Google’s Inception v3 with varying amount of images and their classes. The model was able to classify 15 different classes with 99% accuracy with just 50 training images per class. The datasets used were Animals on the Web, Animals with Attributes 2 and NIH Chest X-ray images. ...learn more
Project status: Concept
Overview / Usage
https://github.com/gyani91/Machine-Learning/blob/master/Transfer%20Learning/Code%20Documentation.pdf
Methodology / Approach
https://github.com/gyani91/Machine-Learning/blob/master/Transfer%20Learning/Report.pdf
Technologies Used
https://github.com/gyani91/Machine-Learning/blob/master/Transfer%20Learning/Report.pdf
Repository
https://github.com/gyani91/Machine-Learning/tree/master/Transfer%20Learning