Machine Learning Frameworks
I attended a webinar by this company called Databricks, and I have to be honest, I was a bit lost in some parts. But I made this table that helps me get an overview of machine learning frameworks:
Machine Learning | Deep Learning | Supporting Libraries | Serving and Monitoring |
---|---|---|---|
Scikit-learn | TensorFlow | Python | MLeap |
Spark MLlib | Keras | R | TF Serving |
H20 | Caffe | Anaconda | Cassandra |
mlpack | PyTorch | Numpy | Redis |
Mahout | Theano | Scipy | TensorBoard |
BigDL | pandas | ||
SparkDL | Matplotlib | ||
PyViz |
I am already starting to get comfortable with a few of them, and lists like this help me keep in mind all the things I still have to learn.
I have started reading some articles in Medium about data science. It is a terrific publication. I like it a lot, and I have just finished reading an article on activation functions for neural networks, and they make a lot more sense to me now.
Finally, I have decided to enroll in the Professional Certificate in Data Science offered by HarvardX on edX. It is a nine-course series, and the cool thing is that it is in R, so it will complement what I have been learning in Python.
I have been researching different courses for my next steps, and I think it will help me get some formal certificates on top of the other classes I am taking. It is amazing how many quality courses by top universities are available in edX and Coursera. If you add that MIT has its whole curriculum available for free online, it is mind-blowing how accessible all this knowledge is nowadays.
One has to be thankful to be living in this information age.