Convolutional Neural Networks
The course is part of this learning path
In this course, discover convolutions and the convolutional neural networks involved in Data and Machine Learning. Introducing the concept of tensor, which is essential for everything that follows.
Learn to apply the right kind of data such as images. Images store their information in pixels, but you will discover that it is not the value of each pixel that matters.
- Understand how convolutional neural networks are essential to the fundamentals of Data and Machine Learning.
- It is recommended to complete the Introduction to Data and Machine Learning course before starting.
Hey guys, welcome back. This is the walkthrough of exercise two in section six. So exercise two is pushing the limits of what your laptop is able to do. In fact, we'll load a dataset that requires enactor that is too big for your laptop to train. And so this will pave the road for the next chapter on GPUs and the cloud. So the idea here is to also build an image recognizer but that is able to recognize not digits but pictures in different categories. Particular, we have 10 categories. Airplane, automobile, bird, cat, etc. And we want to build an image recognition system that is capable of recognizing these.
These are parts of the Cifar 10 dataset and the link is provided here to the dataset which is a very famous dataset in the image recognition that has been used for many benchmarks and so it's good that you get to know it. And as the explanation and as this said, in this exercise, we will essentially see that we need a GPU. So here's what you have to do. Load the cifar dataset and we've actually provided the import folium. Display a few images, see how hard it is to recognize an object, these images have a very small resolution as you will see. So it's actually not easy to spot here for human to understand what's displayed the image. Check the size and the scale of X_train.
Check the shape of Y_train and perform the usual reshaping and rescaling. And then build the model with the following architectures. So convolutional, convolutional, maxpool, convolutional, convolutional, maxpool, flatten, dense, output. And notice that here, we are applying two convolutional layers on stake, on top of one another before we apply the maxpool. Also make sure to choose appropriate activation functions and the parameters for each of the convolutional layers. So compile the model, check the number of parameters and try the training. You will soon see that the model will take forever to train and that's actually expected. So once you're getting to the training step and the model, execute the training, just stop the training and move to the next section. Thank you for watching and good luck.
About the Author
I am a Data Science consultant and trainer. With Catalit I help companies acquire skills and knowledge in data science and harness machine learning and deep learning to reach their goals. With Data Weekends I train people in machine learning, deep learning and big data analytics. I served as lead instructor in Data Science at General Assembly and The Data Incubator and I was Chief Data Officer and co-founder at Spire, a Y-Combinator-backed startup that invented the first consumer wearable device capable of continuously tracking respiration and activity. I earned a joint PhD in biophysics at University of Padua and Université de Paris VI and graduated from Singularity University summer program of 2011.