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

It helps speed up training by only using important data examples

It lets us enlarge image data if they are too small

It lets us simulate additional data examples, which is useful when our dataset is small

It increases the resolution of images so the model has an easier time training

To use many small convolution layers for improved model accuracy

To reduce the number of parameters used by the model's convolution layers

The fire module squeezes and expands the data to reduce noisy features

What is the main purpose of the fire module?

What is the main purpose of the fire module?

It uses no parameters, which reduces overfitting compared to a fully-connected layer

It significantly speeds up model training

It is necessary in order for a CNN to classify multi-channel image data

Which of the following is a reason to use global average pooling at the end of a CNN?

Which of the following is a reason to use global average pooling at the end of a CNN?

It saves a significant amount of space, especially when there are many data classes

It can improve a model's accuracy on outside datasets

It helps a model converge to the optimal weight settings

There are no benefits to using sparse versus dense representation

What is a benefit to sparse representation of labels?

What is a benefit to sparse representation of labels?

What you'll learn from this course

Image Processing

CNN

SqueezeNet

ResNet

Quiz