Teaching yourself deep learning is a long and arduous process.
Knowing where to start, however, can help a lot in softening the learning curve.
Most books on deep learning require a basic knowledge ofmachine learningconcepts and algorithms.
Trasks book teaches you the fundamentals of deep learning without any prerequisites aside from basic math and programming skills.
In contrast,Grokking Deep Learningteaches you deep learning by building everything from scratch, line by line.
40% off TNW Conference!
Trask takes you through the basics of linear transformations, the main computation done by an artificial neuron.
You then implement the artificial neuron in plain Python code, without using any special libraries.
But writing everything in vanilla Python is excellent for learning the ins and outs of deep learning.
Youll then measure the prediction error and apply gradient descent to tune the neurons weight in the right direction.
Again, youll implement everything yourself with the help of Numpy functions.
Youll also learn to compute gradients and propagate errors through layers to spread corrections across different neurons.
The book features some popular regularization techniques such as early stopping and dropout.
If you havent, youll have a much easier time getting comfortable with those libraries in the future.
Code library is a bit cumbersome
Not everything aboutGrokking Deep Learningis perfect.
And in this area, Trask could have done a much better job.
TheGitHub repositoryofGrokking Deep Learningis rich with Jupyter Notebook files for every chapter.
Jupyter Notebook is an excellent tool for learning Python machine learning and deep learning.
Some ofGrokking Deep Learnings notebooks are composed of very large cells with big chunks of uncommented code.
Also, Trask has written the code in Python 2.7.
Deep learning isnt a magic wand that can solve every AI problem.
The point is, youll need a full arsenal of tools and techniques to solve AI problems.
Hopefully,Grokking Deep Learningwill help get you started on the path to acquiring those tools.
Where do you go from here?
You should also deepen your knowledge of other machine learning algorithms and techniques.
Two of my favorite books areHands-on Machine LearningandPython Machine Learning.
The AI universe is vast and quickly expanding, and there is a lot to learn.
If this is your first book on deep learning, then this is the beginning of an amazing journey.
you’re able to read the original articlehere.
