Machine Learning using Python requires knowledge of NumPy and Pandas. Since this discussion series uses Jupyter notebook, a knowledge of Jupyter notebook and Markdown will also be helpful.
A good understanding of Numpy will be accomplished if you go through this attached File.
A good understanding of Python Pandas Dataframe is also required in order to learn machine learning using Python. In this example we will use only python Version 3.
All python packages can be installed through one of the package installation packages such as 'pp3'.
A good understating of TensorFlow is also required to learn machine learning.
Logistic Regression on Titanic Data
Logistic Regression on Bank Marketing Data
We have used online materials extensively during this discussion series. Especially, those materials from Google Machine Learning Crash Course site. The materials are Apache 2.0 licensed and therefor allowed to modify for the discussion purpose.
Google Machine Learning Crash Course
1. Quick Introduction to Pandas
2. First Steps with TensorFlow
3. Synthetic Features and Outliers
4. Validation
1. Linear Regression with One Variable
2. Linear Regression Cost Function
5. Logistic Regresstion Cost Function
REFERENCES:
The Discussion series has used many external references to understand machine learning. Some of the Internet sites are given below. Please keep in mind Internet sites often disappear with time . However, some of the sites have very useful information to advance machine learning knowledge if they are reachable.
2. Stanford Machine Learning Course Notes
3. A good compilation of references and discussions are available at this ml-cheat-sheet site .
4.