Reach for “Good Enough”
Looking at my plans for this past week, I was pretty ambitious in what I thought I’d accomplish. And while I technically started what I said I would, we’ll have to use a very generous definition of the word start to make it work. Sometimes learning machine learning feels overwhelming to me - there’s so much involved! From matrix math and linear algebra and calculus to programming in python to applying everything to machine and deep learning projects.
It’s a lot.
I definitely fall into the trap of “I have to take these 3,098,238 courses first, then I’ll be ready for my first deep learning project!” And everything I know about teaching and education and learning tells me this is absolutely the wrong approach. So instead of trying to perfect my knowledge base before trying something, I’m going to start reaching for “good enough”.
Reaching for “good enough”: me + Adobe
I spent the bulk of my week focusing on the Adobe Creative Suite for a work project, and dove headfirst into Lightroom, Photoshop, Illustrator, Audition, and Premiere Pro in order to get this project across the finish line. It was wild and a ton of fun and constant learning on the fly.
I didn’t create any award winning material, but I’m proud of the video I made and what I learned to make it happen. Because based on my constraints - time, what I knew, what I could learn, and what I was trying to accomplish - this video is good enough. What’s more, I’m excited about creating another video, and another video after that, and maybe this a new hobby worth exploring.
Last week’s learning recap
Things went a little better this week compared to last, and some things in Python are starting to click (finally!).
Finished! Whirlwind Tour of Python (solo).
Everything I said about this book last week is still true - this is the best introduction to Python I’ve had, and I’d recommend it without reservation to anyone who is familiar in one (or more!) programming languages and looking to learn Python.
Started! Python Data Science Handbook (solo)
I haven’t made it far into the text, but it’s nice to feel like I’m making progress on something that will be immediately applicable to my machine learning journey!
Started! Hands-On Machine Learning with Scikit-Learn, Keras, and Tensorflow (small group)
The original plan was to make it through the first two chapters, and then spend an online session doing the coding exercises in chapter 2, but between a work project and submitting our Data Science in Education manuscript to the publisher, I couldn’t get through chapter 2. So we’re trying again this week!
Continuing fast.ai Practical Deep Learning For Coders course (large online group)
I appreciate the way that the content in the course is designed, and that the creators have taken the “whole learning” approach from Perkins with regards to curriculum design, but the delivery is the same as most online courses I’ve tried: the instructor reads to you for the duration of the class. I’m not sure that I’m getting anything out of attending the lecture that I don’t get from reading the book and completing the exercises, but I do like having the group structure to help with deadlines.
This week’s learning agenda
A lot of the same things from last week!
- Week four of fast.ai Practical Deep Learning For Coders course, along with working through the coding exercises and end of chapter questions. One thing I want to commit to is spending time going through the documentation while doing the coding exercises, so that I can better understand what’s happening with my code.
- Chapter 2 of Hands-On Machine Learning with Scikit-Learn, Keras, and Tensorflow
- Continue with Python Data Science Handbook