Gain insights into image data processing and CNNs for image classification using TensorFlow. Delve into CNN architectures and their applications, requiring Python and TensorFlow knowledge.

irml.tar.gz

image-rec

If you want to dive into the technology behind computer vision and self driving cars, this is where to start.

In this course you'll learn how to process data from image files and create convolutional neural networks (CNNs) to classify different types of images. After completing this course, you will have a solid understanding of why CNNs work and which CNN architectures to use for different tasks.

The code for this course is built around the TensorFlow framework, one of the premier frameworks for industry machine learning, and the Python pandas library for data analysis. Knowledge of Python and TensorFlow are prerequisites. 

This course was created by AdaptiLab, a company specializing in evaluating, sourcing, and upskilling enterprise machine learning talent. It is built in collaboration with industry machine learning experts from Google, Microsoft, Amazon, and Apple.

Image Recognition with Machine Learning

<h2 class="section-title">Chapter Goals:</h2>
<ul>
	<li>Learn about data augmentation and its purpose</li>
	<li>Apply data augmentation to image data</li>
	<li>Standardize the image data</li>
</ul>


<h2>A. Data augmentation</h2>

<div id="section1_l2" class="collapse show">
<p><i>Data augmentation</i> is the process of artificially enlarging a dataset through <i>image transformations</i>. No matter how good a model is, it cannot perform at its fullest potential unless we have a large enough dataset. Although CIFAR-10 contains 50,000 training images, it would benefit our model if it were trained on more data. Rather than going through the process of obtaining new images and resizing them, we can instead perform transformations on the images (e.g. rotating or cropping) to manufacture "new" image data.</p>
</div>


<p>Using data augmentation, we create a variation of each image in a batch of input data. Since we train our model on hundreds of epochs (full run-throughs) of the CIFAR-10 dataset, data augmentation constructs hundreds of variations for each image in the training set.</p>

<h2>Chapter Goals:</h2>
<ul>
	<li>Learn about data augmentation and its purpose</li>
	<li>Apply data augmentation to image data</li>
	<li>Standardize the image data</li>
</ul>


<h2>A. Data augmentation</h2>

<div id="section1_l2">
<p><i>Data augmentation</i> is the process of artificially enlarging a dataset through <i>image transformations</i>. No matter how good a model is, it cannot perform at its fullest potential unless we have a large enough dataset. Although CIFAR-10 contains 50,000 training images, it would benefit our model if it were trained on more data. Rather than going through the process of obtaining new images and resizing them, we can instead perform transformations on the images (e.g. rotating or cropping) to manufacture "new" image data.</p>
</div>


Discover how data augmentation works and its benefits for training a CNN.

What you'll learn from this course

Image Processing

CNN

SqueezeNet

ResNet

Augmentation

Chapter Goals:

A. Data augmentation