This guide is designed to be the quickest way to get Python, Jupyter Notebook, and several data science related packages up and running on a Windows computer. This free and open software can do some amazing things. No prior experience is required.
1) Install Python 3
a. Download the latest stable release at:
b. The first link under “Stable Release” was Python 3.9.5 - May 3, 2021 as of May 24th, 2021 so click on that or perhaps a newer one.
c. Scroll to the bottom. Under “Files”, click Windows installer (64-bit)
d. After the download completes, double click the “python-3.9.5-amd64.exe”.
e. Check the “Add Python 3.8 to PATH” checkbox.
f. Click Install Now → Yes → “Setup was successful” should appear → Close
2) Install Anaconda
a. Download the data science toolkit from Anaconda at: https://www.anaconda.com/products/individual#windows
b. Click Download → Click 64-Bit Graphical Installer. The download should start immediately.
c. After the download completes, go to your Downloads folder double click “Anaconda3-2021.05-Windows-x86_64.exe”.
d. Click Next→ I Agree → Next → Next → Install → Next → Next → Uncheck both box checkboxes. → Finish
3) Install Jupyter Notebook
a. Hit the Windows key which is by the “Z’ key → Type “anaconda n” → Click “Enter” → this will launch Anaconda Navigator.
b. Click the “Launch” button under “jupyter Notebook 6.3.0”.
c. This should launch “localhost:8888/tree” page on your web browser.
4) Install Git Bash
a. Go to: https://git-scm.com/downloads
b. Click the “Windows” logo. The download will begin automatically.
c. After the download completes, double click “Git-2.31.1-64-bit.exe”.
d. Click “Yes” → Next 15 times → Install → Uncheck View Release Notes → Finish.
Hit the Windows key which is by the “Z’ key → Type “gitb”→ Click “Enter”.
e. Type “cd doc” → Click “Tab” → “cd Documents/” should appear → Click “Enter”.
f. To make sure that “pip” the standard package manager for Python is working, type
pip -V →
output like “pip 21.1.1 from c:\user\... (python 3.9)” should appear.
g. Make sure pip is up to date by typing
py -m pip install --upgrade pip.
5) Create a virtual environment. This enables this project to have its own dependencies.
python -m venv venv
b. Activate the new virtual environment with
“(venv)” should be visible three lines up from the bottom left.
c. Install Python packages needed to perform data science.
pip install matplotlib && pip install pandas && pip install seaborn && pip install seaborn && openpyxl
d. Test the install by running this finances tracker example from my github page.
git clone https://github.com/timg1234/Personal-Finances-Tracker
cd pers→ Click “Tab” → Enter.
f. Go back to your web browser (probably Chrome) that's running Jupyter Notebook → click “Documents” → click “Personal-Finances-Tracker” → click “Finances Tracker.ipnyb” → you should see a GIF of a hilarious character from one of the best comedies of all time.
g. To the right of “File, Edit...” click “Kernel” → “Restart and Run All” → click the red “Restart and Run All Cells”.
And boom, a timeseries analysis of some fictional financial date should appear. Note that the time you ran this at should appear in the filename saved to the PDF. If you repeat step d. a moment later, you’ll see that the timestamp to the filename has been updated.