My Second Project in Machine Learning Engineering Nanodegree

Deep Learning

Project: Object Classification

Table Of Contents:

• Description
  
  • About the project
    
  • What needs to be done
    
• Data
  
  • Files
    
  • Dataset file
    
• Loading Project
  
  • Requirements
    
  • Execution
    
• Conclusion
  
  • Evaluation
    
  • Results

Description

About the project

Classifying 'Objects' from CIFAR-10 dataset which consists of 60000 32x32 color images in 10 classes, with 6000 images per class. There are 50000 training images and 10000 test images. The dataset is divided into five training batches and one test batch, each with 10000 images. The test batch contains exactly 1000 randomly-selected images from each class. The training batches contain the remaining images in random order, but some training batches may contain more images from one class than another. Between them, the training batches contain exactly 5000 images from each class.

What needs to be done

The dataset will need to be preprocessed, then train a convolutional neural network on all the samples. Then I'll normalize the images, one-hot encode the labels, build a convolutional layer, max pool layer, and fully connect the layer. At then end, I'll see their predictions on the sample images.

Data

Files

This project contains 3 files and 1 folder:

• report.ipynb: This is the main file where I have performed my work on the project.
• problem_unittests.py: Helper File.
• helper.py: Helper File.
• export/ : Folder containing HTML version file of notebook.
• result.png : This image shows final results of the model.

Template code is provided in the report.ipynb notebook file. While some code has already been implemented to get me started, I will need to implement additional functionality when requested to successfully complete the project.

Dataset file

The dataset I have used is provided by CIFAR-10 dataset.

Loading Project

Requirements

This project requires Python 3.6 and the following Python libraries installed:

• Python 3.6.5
• NumPy
• TensorFlow

You will also need to have software installed to run and execute a Jupyter Notebook

If you do not have Python installed yet, it is highly recommended that you install the Anaconda distribution of Python, which already has the above packages and more included.

Execution

In a terminal or command window, navigate to the top-level project directory Image_Classification/ (that contains this README) and run one of the following commands:

ipython notebook image_classification.ipynb

or

jupyter notebook image_classification.ipynb

or if you have 'Jupyter Lab' installed

jupyter lab

This will open the Jupyter/iPython Notebook software and project file in your browser.

Conclusion

Evaluation

My project was reviewed by a Udacity reviewer against the Image Classification project rubric. All criteria found in the rubric must be meeting specifications for me to pass.

Results

My Project Review by an Udacity Reviewer