Just a few weeks ago, Enthought released this year's SciPy tutorials on machine learning with scikit-learn, taught by Andreas Muller and Alexandre Gram.
This day-long tutorial was divided into two sessions, a morning and an afternoon session. There's a video for each session and these videos are about 3 hours and 10 minutes long.
Here's part of what's covered in the first session:
- basic concepts of ML and data representation (types of learning in ML)
- preprocessing
- dimensionality reduction
- clustering
Algorithms covered include: nearest-neighbors, linear models, tree-based models, SVMs, and neural networks.
The afternoon session covers optimization of models through hyperparameter tuning and prevention of overfitting - to name just two. It also gets into model complexity, building pipelines, how to deal with large datasets, and a few more advanced topics.
Please see the following video for the session. You can continue with the second session which is linked in the right sidebar of related videos. There's no excuse in not getting better at something - you have everything you need (the materials), and it's free. So, you need to put in the work. Good luck.
To stay in touch with me, follow 
Cristi Vlad, Self-Experimenter and Author