Hi,

Yesterday I got an email from Mateo, a computer vision + deep learning researcher that was so funny (and 100% accurate) that I had to share it with you.

Mateo writes:

~~~

As a deep learning researcher, I can tell you that detailed code/implementations and experiment journals are invaluable to developing a successful project.

In countless previous occasions I've found that books, publications, and sample codes gloss over critical steps and leave me to "draw the rest of the %$@# owl."

Adrian, your deep learning book contains the most detailed code, implementations, and experiment logs I have ever seen. Thank you for putting it together, it's been worth every penny.

~~~

If you're not familiar with the "trying to draw an owl" reference Mateo is making, it comes from this popular internet meme:

Image

I have blurred out the curse word in the image, but I think you get the picture.

Too often you're confronted with books, courses, and tutorials that gloss over critical steps — this is especially true for machine learning and deep learning tutorials.

Similar to the drawing in the image above, you find yourself presented with the very first step where you're instructed to open a code editor and import a few packages and load an image from disk.

The first step makes total sense and you're following along just fine.

You may even feel excited and energized, like you're really building up a head of steam...

...only to find that the next step in the tutorial:
  • Attempts to review a 50 line code block in just two sentences
  • Initializes objects and classes you've never heard of
  • Calls functions that you don't understand what they are doing (or why they are important)
  • Discusses zero theory, leaving you without the knowledge to transfer this example to your own projects or research
In essence, these guides skip to the end result.

But what about the in-between steps? It seems like we jumped from immediately from step A to step Z. Aren't those other steps important too?

Your motivation hits a brick wall, stopped dead in its tracks, as you, frustrated, scratch your head.

If you've ever felt this way (I know I have) I would really suggest you take a look at the table of contents and sample chapters of my book, Deep Learning for Computer Vision with Python:

>> Click here to see the table of contents + sample chapters

Once you dive in you'll see that I don't skip steps.

I don't expect you to "draw two circles" and then "build the rest of the Convolutional Neural Network, train it, and obtain high accuracy".

Instead, I guide you through each step, carefully taking the time to dissect not only the theory but the implementation as well.

If you're anything like Mateo, you'll appreciate the time, effort, and energy I've put into making this the most complete and comprehensive deep learning book online.

Cheers,

Adrian Rosebrock
Chief PyImageSearcher

P.S. Ready to join me, Mateo, and thousands of other PyImageSearch readers in deep learning mastery?

Just click here to grab your copy of the book — I'll be with you every step of the way (and I won't skip any steps either).