#7 Week of the Year
For the present ending the Deep Learning journey and going back to the roots — whatever that means.
I never guess. It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts.
- Sir Arthur Conan Doyle, Author of Sherlock Holmes stories
This weeks summary begins with a quote from one of the most famous authors of British literature. I didn’t choose the quote, because it’s from a famous author, but rather because of it’s meaning.
Data becomes more and more important in our world. Not only in the business world we are collecting and analyzing data to predict the future. We also do it in politics, healthcare and sports to name a few.
That’s why I adjusted my learning path a little more.
From Deep Learning to Data Science
My original goal was to learn more about Deep Learning. Therefore I gained basic knowledge in Data Science and Machine Learning to be able to understand later, what Deep Learning really is. I used Python and TensorFlow to create own models helping me to decode the received information.
Quickly I struggled on it, because there where to many knowledge gaps. Nevertheless I kept going and finished the corresponding courses at CognitiveClass.
After these experiences I’m aware of the complexity of this topic. So, how to go about it? Answer: Back to the roots!
I started the Data Science Fundamentals learning path at CogntiveClass to learn more of the basic concepts of data science. Additionally I will extend my R knowledge to be able to do data science stuff (data mining, transformation, processing, etc.) not only in Python.
Furthermore I found some nice podcasts covering these topics. I will add them to the list at the end of the article.
Docker: Shipping containers around
Recently I listened to the CodeNewbie episode “What’s a container?” and wanted immediately to learn more about the container technology in general and Docker in specific.
After a quick research on the topic and reading the Wikipedia article about it, I found a nice course at IBM developerWorks. This course covers all the basic knowledge you need to have to understand the concept of containers and how you can use them for your projects.
With an aroused interest I started planning how to refactor and move older projects to containers. Dealing with the “Doesn’t work on my machine”-issue will be from now part of the past.
Bash-Scripting on OSX
In the past weeks I kept going thinking about things, I have to deal with and which I could automated using bash-scripts.
While doing and completing courses online, I had to download resources for offline studying. One of the resources were Jupyter Notebooks. I downloaded a bunch of the to realize, that there was always the .json extension at the end of the file (right after .ipynb). The easiest thing to get the notebook running is simply to remove the extension.
My goal was to remove the .json extension from all the notebooks before starting jupyter notebook. Therefore I wrote a little bash-script. But I wanted to have them in PDF-format as well, so that I could easily print them. So I extended my script to be able to add a ‘- -pdf’ flag converting all notebooks in the directory and sub-directories to PDF-format.
Script:
I only have one issue I wasn’t able to fix up to now. Once in a while it happens, that I get an error, that the notebook.tex file is not existing. This file is generated through the converting process by the nbconvert command.
Home automation and working on sensors
Over the past week I went on working with sensors and their data. I made a couple of breakthroughs and started creating persistence strategies.
Tomorrow I’m going to start the persistence process and will see how to go on on this topic.
I didn’t gain new crucial knowledge and had to deal with minor not appreciable issues. That’s why I’m not going into detail what I’ve done.
List of worth reading articles and podcasts
On my way to work, the way back home or in other situations I’m reading a lot of articles about the topics I’m dealing with as well as listening to podcasts. In the list beneath I want to share some of them with you — or at least those I regained.
Podcasts:
Articles:
- 7 Keys to Winning the Morning
- Python Tuples Tutorial
- A brief tour of Python 3.7 data classes
- Jupyter Notebook Tricks for Data Science that Enhance your efficiency
For next weeks summary I’m going to save the links to the articles I’ve read and which were worth sharing. Wishing you a week full of coding — Keep coding!
