Friend is looking for a quick start into tensorflow / deeplearning. Below’s a quick start if you’re in a similar situation - 

  1. Learn Git (10 hrs)- If you want to collaborate, you need to know this, period. (was that redundant?)
  2. Setup Google Datalab (10 hrs) - 
    1. You’ll spend 10 hrs trying to configure your computer to run things  locally only to realize that you’ll have to run this code in the cloud anyway. 
    2. Google has a native(ish) integration with Tensorflow.
    3. This is a really good place to explore code since it’s Jupyter + cloud.
  3. Go through the tutorials (100 hrs)
  4. Go through some videos and the “live” programming sessions. (100 hrs)
  5. Explore other people’s code (10 hrs)

The time estimates are generous and if you’re spending longer than that on any given section, then ask for help :)