Skip to content

Latest commit

 

History

History
48 lines (42 loc) · 2.94 KB

README.md

File metadata and controls

48 lines (42 loc) · 2.94 KB

Numpy and matplotlib tutorial for beginners using a breastcancer dataset.

Running this tutorial online:

You can Launch this tutorial on Binder or Google Colab. The slide-show will work on binder, but not colab. Press Alt+r to toggle slide-show.

Running this tutorial on your own computer:

You need to have Python installed with numpy, matplotlib and pandas libraries. https://www.anaconda.com/ Python distribution is easy to install and bundles common libraries for scientific computing.

  1. Click the Clone or download button.
  2. Select Download ZIP.
  3. Unzip the downloaded zip file into a folder.
  4. Open a terminal (command prompt/anaconda prompt on Windows) and go to this folder by entering cd folder-path command).
  5. Start jupyter notebook (enter jupyter notebook). This will open a browser window showing the contents of the folder.
  6. Click the Wisconsin_breast_cancer_data.ipynb file and this will open the notebook in a new window or tab of the browser.
  7. In order to run it as slide-show you must install https://rise.readthedocs.io module first. Press Alt+r to toggle slide-show.

Summary

This tutorial explores basic usage of numpy and matplotlib using this publicly available dataset: http://archive.ics.uci.edu/ml/machine-learning-databases/breast-cancer-wisconsin.

  • Introduction to jupyter including running it on the cloud.
  • Import python modules.
  • numpy basics
    • array creation, arithmetic, indexing, slicing, reshaping.
    • heterogeneous arrays (structured/record arrays with named fields).
  • load data from csv (text) file.
    • (advanced) use requests library to retrieve data from the Internet.
    • (advanced) use StringIO to read from string containing the csv data.
    • (basic) specify field names and data types using numpy dtype.
    • (basic) specify missing value handling when loading data
    • (basic) check columns for missing values.
  • Pandas
    • (basic) load data using Pandas.
    • (basic) accessing rows and columns in Pandas dataframe.
    • (basic) do simple boxplots of dataframe columns. -matplotlib
    • basic plotting
      • line plot
      • scatter plot
      • histogram
      • box plot
    • legend
    • subplots
  • list of other useful modules

History:

It was originally the first part of a two part workshop on the basics of numpy, matplotlib and pandas for data science. The first version used in the live workshop can be downloaded as an archive here: https://github.com/subhacom/np_tut_breastcancer/releases/tag/v0.1. This notebook may be updated occasionally.

The second part of the workshop focuses on pandas for data science and is available here: https://github.com/bballew/pandas_tutorial.