My Machine Learning Workflow

• Import Libraries
• Import Data
• Look at top 10 values to see how data looks like
• Data Frame Profiling
  • Looking at shape
  • Talk to domain expert
  • Count NA values in each column and replacing them if needed
  • Data cleaning
  • Describing numerical variables
  • Checking for outliers using boxplots and removing them
• Machine Learning
  • Correlation Heatmap
  • Feature Engineering (https://towardsdatascience.com/feature-selection-techniques-in-machine-learning-with-python-f24e7da3f36e)
  • Checking assumptions of the model, thinking about parameters
  • Applying ML algorithm
  • Drawing real and predicted outputs for graphical understanding
  • Improving the model and sharing outputs