This repository is dedicated to helping newcomers to data science who desire a basic primer to get swiftly situated and practicing data science. I don't intend for this resource to be comprehensive. A much more thorough primer covering basic statistics, probability, linear algebra, machine learning and more will be covered in a MOOC format that I intend to release upon near completion. Please stay tuned for that.
If you come out of this tutorial thinking that you now have a good feeling for how to tackle a data science problem and can abstractly describe a pipeline to do so, then I feel this tutorial has satisfied its purpose.
Please get started by cloning this repository locally to your workstation. All of the working files are Jupyter Notebook files denoted by the .ipynb extension. If you are unfamiliar with Jupyter Notebooks, please check out this YouTube video. The notebooks work together with the data files housed adjacent to the repository folder and entitled "data_files". Rest assured, they are very light-weight files (under 5MB) so feel free to clone the entire repository.
Please note that there is a supporting blog on Medium that goes in concert with the Jupyter Notebooks on the Github repository.
- Python 2.7x Installed (Note 3.x posesses similar syntax)
- Working knowledge of Python
- Jupyter Notebook
I also recommend having a virtual environment for this tutorial. Please consider using Virtualenv or Anaconda.
The tutorial moves swiftly as it assumes working knowledge of the Python programming language, Pandas and NumPy. Please consider these resources to fill in some of the deliberate omissions in this tutorial.
- ipykernel==4.5.2
- ipython==5.1.0
- jupyter==1.0.0
- matplotlib==1.5.3
- nbconvert==5.0.0
- numpy==1.11.3
- pandas==0.19.2
- pickleshare==0.7.4
- pyparsing==2.1.10
- python-dateutil==2.6.0
- requests==2.12.4
- scikit-learn==0.18.1
- scipy==0.18.1
- seaborn==0.7.1
- statsmodels==0.6.1
If your virtual environment is through Anaconda, then use conda install command to get the necessary packages. If you use Virtualenv, then you can use the pip install command. Documentation for both package managers are robust, so please refer to those as needed.
I intend to create a separate Python 3.x version of this repository. Note that syntax difference is minimal for this tutorial.
- D.H. Kim - Initial Work - dhexonian
This project is free to distribute and utilize with good intent.
Please consider following me on Twitter @dhexonian and Medium @dhexonian. I'd love to hear about how I can best serve the learning community.
To my family for being patient and loving.