Welcome to the landing page of the Contently Math Class! The hope here is that we can all have fun while diving into some of the knowledge and technique that allows for thinking quantitatively.
To achieve this, we will make use of the Python programming language. Please note that this is not meant to be a "how to write Python" class. I.e. our focus will be not on the code but on the numbers, and we will generally write simplistic representations of our desired logic. That said, you will gain exposure to writing analytical Python through these courses.
Each lesson's content will be delivered in the form of a Jupyter notebook. This is a browser-based, interactive coding tool that allows you to keep code, charts, and text in the same place. GitHub will be used for distributing these notebooks - you can in fact view the notebooks right through GitHub but you'll have to run Jupyter locally if you wish to interact with them. If you have any questions on how to use Jupyter, please don't hesitate to reach out at any time. I imagine you will pick this up pretty quickly.
That said, it's hard to beat a good 'ol fashioned pen and paper for math, and if you'd like to participate in this way you should feel free to do so! This will work for many, though not all, of the classes. You're welcome to try to follow along in another language, too. Note that this would likely leave you on your own with respect to implementation details, which could become problematic as we get farther along. But you should feel free to experiment as you wish - you can always come back to working in Python. Please see here for a list of resources to help you further your Python, and feel free to update it with your thoughts and recommendations!
If you do wish to use Python, I would recommend using at least version 3.6. You will also need to install the jupyter package. Additionally, the following analytical libraries would be helpful to have ready to go:
numpyscipymatplotlibpandassklearnstatsmodelsrequests(this is for HTTP - always handy to have installed)
The simplest approach for folks who don't already have a good Python setup is likely to download and install the Anaconda Python distribution, which will include all of these packages and also many others. You can download that here.
Otherwise, once Python 3.6+ is installed, you can either run pip install -r requirements.txt from within this directory or go ahead and just individually install each of the above packages.
If you are running Docker, you can use the start command from the terminal to simply load the notebook without installing dependencies or python locally.
$ ./startAfter the Docker images are set up, you will see a url listed at the end similar to this:
Copy/paste this URL into your browser when you connect for the first time,
to login with a token:
http://(acd90c62cec2 or 127.0.0.1):8888/?token=9647b38f8622f7f33ccf970ecee3a0177e37d3e42ccb2932
Your url is most likely:
http://localhost:8888/?token=<that token above>
- Numbers and Operators
- Basic Probability
- Functions
- Calculus I
- Calculus II
- Linear Algebra
- Functions II
- Calculus III
- Decision Trees
- Probability & Statistics II
- Linear Regression
- Differential Equations
- Logistic Regression
We have a few months' worth of topics described above. If the class is working well - i.e. folks are learning and enjoying - we can continue on with some additional topics, such as Autoregression (regressing data against itself, which can prove quite effective for modeling time series data), Robust Statistics (a sub-field of Statistics that provides techniques that hold up in face of outliers), and (what I'd call) Transformations (the restructuring of data to make it more suited for certain types of analyses).
Furthermore, pending the proper interest as well as the completion of our fundamentals review, we could let this evolve towards something like an "intro to data science" class.