Exploring a Dataset before getting started or preparing it for a machine learning model is really important because we should know what the data is telling and what all features and target column has. Generally, we start by finding out the relationship between different columns, creating different types of visualizations, finding patterns in data, etc. According to a study Exploratory Data Analysis consumes around 40% of the total project time, What if I say that it can be reduced to 10%? This means now you can focus more on building a robust and highly accurate Machine/Deep Learning Model without wasting much of your time in EDA. Sweetviz is an open-source python library that creates an EDA report in just a single line of code. Yes, you read it correctly, in just a single line of code, you can analyze data, visualize the patterns and associations of different columns in the dataset.

Machine Learning is 80% preprocessing and 20% model making. You must have heard this phrase if you have ever encountered a senior Kaggle data scientist or machine learning engineer. The fact is that this is a true phrase. In a real-world data science project, data preprocessing is one of the most important things, and it is one of the common factors of success of a model, i.e., if there is correct data preprocessing and feature engineering, that model is more likely to produce noticeably better results as compared to a model for which data is not well preprocessed. There are 4 main important steps for the preprocessing of data.