Skip to content

Applying neural networks and transfer learning techniques with Tensorflow to classify flower images and build a command-line app

Notifications You must be signed in to change notification settings

rynparin/Image-Classifier-Tensorflow

Repository files navigation

Image-Classifier-Tensorflow

In this project I apply neural networks and transfer learning techniques to classify flower images and build command line app

Intro to Machine Learning - TensorFlow Project

Project code for Udacity's Intro to Machine Learning with TensorFlow Nanodegree program.

Data

Oxford of 102 flower Image

Contents

  • Project_Image_Classifier_Project.ipynb: Jupyter notebook showing all the steps to create a classifier model
  • Project_Image_Classifier_Project.html: HTML version of the jupyter notebook
  • ryn_model.h5: Classifier model
  • predict.py: Python application that uses our classifier model to predict flower type

Command line app

In predict.py I build a command line app that can use to predict image which you can choose image, model, class_name

Usage

> predict.py [image_path] [model_path] --top_k [number_of_classes] --category_names [json_file_of_class_names]

Sample:

> predict.py ./myflower.jpg my_model --top_k 3
> predict.py ./myflower.jpg my_model --category_names map.json

About

Applying neural networks and transfer learning techniques with Tensorflow to classify flower images and build a command-line app

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published