Step by Step with Python: A Beginner's Journey through Data Reading, Wrangling, and Machine Learning


Mar 20, 2024 | By Ananya Chakraborty

Step by Step with Python: A Beginner's Journey through Data Reading, Wrangling, and Machine Learning

In today's data-driven world, understanding Python's role in data collection, data reading, data wrangling, and machine learning is crucial. Whether you are a complete beginner or a seasoned data analyst, knowing how to perform and automate tasks in Python will help you improve your data science skills. In today's article, we will discuss the five essential tasks of data science that you can perform easily with Python.

Essential Tasks of Data Science Using Python:

  1. ๐‘๐ž๐š๐๐ข๐ง๐  ๐ƒ๐š๐ญ๐š ๐…๐ข๐ฅ๐ž๐ฌ

  2. ๐’๐ฎ๐›๐ฌ๐ž๐ญ๐ญ๐ข๐ง๐  & ๐Œ๐จ๐๐ข๐Ÿ๐ฒ๐ข๐ง๐  ๐ƒ๐š๐ญ๐š

  3. ๐๐ซ๐ž๐ฉ๐ซ๐จ๐œ๐ž๐ฌ๐ฌ๐ข๐ง๐  & ๐€๐ ๐ ๐ซ๐ž๐ ๐š๐ญ๐ข๐ง๐  ๐ƒ๐š๐ญ๐š

  4. ๐๐ฎ๐ข๐ฅ๐๐ข๐ง๐  ๐Œ๐‹ ๐Œ๐จ๐๐ž๐ฅ๐ฌ

  5. ๐•๐ข๐ฌ๐ฎ๐š๐ฅ๐ข๐ณ๐ข๐ง๐  ๐“๐ซ๐ž๐ง๐๐ฌ ๐š๐ง๐ ๐๐š๐ญ๐ญ๐ž๐ซ๐ง๐ฌ

Prepare to embark on your data science journey with Python and dive into a world of learning.

Reading Data Files in Python (Start Here!)

Reading and importing data in the Python environment is the very first step in a typical data science project. Python consists of libraries that help us import data & read the data files in various formats like CSV Files, Text Files, Excel lines, and JSON lines.

Why is it needed?: The very first step in the analysis is to read the data files and understand what the data is all about. While there are other approaches to reading data files, using Python to read and analyze datasets will help us tackle the challenges of importing such data files. Some of the common challenges may be formatting, empty rows, and columns, which can be easily handled using Python.

How to implement it?: Python Libraries like Pandas make reading data files very user-friendly, allowing beginners to read data files in a few lines of code.


import pandas as pd

data_files= pd.read_csv('filename.csv')


Things to Remember:

  • Reading data files is the first step in analyzing data.

  • Ensure that the file path of the data files is correctly mentioned while importing the data in Python Environment.

  • With the help of Python Libraries like Pandas data reading, data manipulation is easier.

  • Be comfortable working with different data file formats and learn to tackle challenges that come with reading other data files.

  • Explore the first couple of rows, data types, and summary statistics to derive valuable insights.

  • Functions like data.head(), data.dtypes and data.describe() will come in very handy during your initial analysis.

Subsetting & Modifying Data (Early Exploration)

After loading the data in your preferred Python notebook, it is time to filter the data and prepare specific subsets of the dataset to focus on the details that are relevant to the project,

Why is it needed?: To eliminate the noise and messiness in the data, tailoring our data as per the requirements helps us concentrate on the crucial aspects of the data,

How to implement it?: This time again, the Python library Pandas has helpful functions that help us filter rows, select specific columns or rows and create new variables for deeper analysis of the dataset.

subset_data1 = data[data['column_name'] > value]

Things to Remember:

  • Filtering data is critical for targeted analysis.

  • Pandas provide easy-to-use functions for exploring the data.

  • Understanding your data's structure helps in effective subsetting.

  • Tools like histograms, scatter plots, and correlation matrices can be valuable at this stage.

Preprocessing & Aggregating Data (Data Preparation)

This stage is crucial for ensuring that your data is ready for analysis. Preprocessing involves cleaning, transforming, and summarizing data.

Why is it needed?: Proper data preparation ensures accuracy, helps uncover key insights, and removes inconsistencies. It's about making your data as clear and relevant as possible.

How to implement it?: Various preprocessing techniques and aggregation methods are available in Python, including handling missing values, transforming variables, and summarizing data.

# Filling missing values
data['column'].fillna(value, inplace=True)
# Aggregating data
summary = data.groupby('category').sum()

Things to Remember:

  • Data preparation is vital for accurate analysis.

  • Understanding the nature of your data and the requirements of your analysis will guide your preprocessing steps.

  • Common tasks include handling missing values, transforming variables, feature engineering, and summarizing data.

  • Visualization tools can help identify outliers and distribution characteristics.

  • Libraries like Pandas and NumPy provide extensive functionality for data cleaning and transformation.

  • Libraries like sklearn.preprocessing provide powerful tools for scaling and transforming features.

Building ML Models (Modeling Stage)

Building machine learning models teaches computers to recognize trends and insights and enable decision-making based on the data. It involves steps like selecting, training, and evaluating algorithms to make predictions or classifications.

Why is it needed?: As mentioned earlier, building machine learning models allow us to make predictions, classify information, or generate new data. For example, we can predict weather or stock prices, classify emails as spam or not, and build a system that recommends products to targeted customers.

How to implement it?: Python offers plenty of useful libraries and tools that simplify the process of building machine learning models. Libraries like Scikit-learn provide a range of beginner-friendly algorithms, from classification to clustering, which aids in building ML models.

from sklearn.linear_model import LinearRegression
model = LinearRegression().fit(X_train, y_train)
predictions = model.predict(X_test)

Things to Remember:

  • Machine learning models provide predictive insights.

  • Scikit-learn is a go-to library for beginners.

  • Model selection depends on the nature of your data and the problem you're solving.

  • Understanding the underlying principles of the chosen algorithm and proper validation techniques is essential for building robust models.

Visualizing Trends and Patterns (Insight Discovery)

Data visualization is the art of presenting data graphically via charts or dashboards. It helps in creating compelling representations of our data, trends, and patterns.

Why is it needed?: To understand any form of data, whether simple or complex, visualizing the data makes it understandable, and we can draw some valuable insights from it. Through visuals and charts like stacked bar graphs, scatter plots, and pie charts, we can present our findings to stakeholders and senior managers and have supporting evidence to reveal hidden patterns.

How to implement it?: Python libraries like Matplotlib and Seaborn make data visualization easily accessible, and users can customize it to their preferences.

import seaborn as sns
sns.scatterplot(x='column1', y='column2', data=data)

Things to Remember:

  • Use data visualization techniques to uncover hidden patterns.

  • Choose data storytelling techniques to connect with the audience. (Learn the art of storytelling with data here)

  • Select the correct type of charts and plot types to present the data properly.(Learn How)

  • Avoid fancy-looking visuals; go for simpler and to-the-point charts that can make your visualizations more effective and engaging.


Embarking on the data science journey with Python is an exciting and rewarding endeavor. By following these five essential steps, you can start reading, wrangling, and rocking machine learning like a pro. Practice and continuous learning are key to success in this exciting field. Whether you're a student, professional, or hobbyist, these steps provide a solid foundation for your data science journey. Happy coding!

Key Takeaways:

  1. Python's Versatility: Python simplifies data reading, wrangling, and machine learning through user-friendly libraries.

  2. Systematic Approach: Five essential tasks guide the data science journey, from reading data to visualizing insights.

  3. Beginner Accessibility: Python's intuitive tools make data science approachable for newcomers.

Data-Driven Decisions: Python enables turning raw data into actionable insights, enhancing decision-making across domains

Share With Friends

8 Must-Have Skills to Get a Data Analyst Job in 2024 5 Ways SQL is Used In The Data Industry