Your cart is empty.
Keep shopping
Data Science with Python: Pandas, ML & Model Deployment
132 students
Data Science with Python: Pandas, ML & Model Deployment
Clean data with Pandas, query SQL, train XGBoost models and deploy them as live APIs with FastAPI in hands-on labs.
Created byYotta Academy
Last updated 10/2026
English
English [Auto],
What you'll learn
- Write clean, modular Python using a professional setup: VS Code, Git and virtual environments
- Clean and reshape messy data with NumPy and Pandas: vectorization, multi-indexing, RegEx and mapping
- Pull real-world data with SQL (CTEs, window functions) and REST APIs, and store it efficiently in Parquet and Feather
- Engineer features and handle imbalanced datasets with SMOTE to build a loan default classifier
- Train and tune Random Forest, XGBoost, LightGBM and CatBoost models using Optuna and GridSearchCV
- Explain model predictions with SHAP values and deploy your model as a live web service with FastAPI or Flask
Explore related topics
Course content
8 sections • 32 lectures • 2h 7m total length
Requirements
- Basic Python Knowledge: Familiarity with variables, loops, and basic functions is recommended.
- A Computer (Windows/Mac/Linux): Capable of installing Anaconda or VS Code (we will walk through the environment setup step-by-step).
- No Advanced Math Required: While we cover statistical foundations, a basic high school math background is completely sufficient.
Description
This course contains the use of artificial intelligence.
Most data science courses stop at syntax. Then you open a real dataset and nothing works.
This course is different. You learn by fixing broken data, pulling live data and shipping real models. Builder to builder.
Who this course is for:
- Students who know basic Python and want a job-ready data science workflow
- Developers moving into data science and machine learning
- Freelancers who want to deliver ML models clients can actually use
- Anyone tired of theory-only courses
Level 1: Data Engineering & Statistical Insights
First, set up like a pro: VS Code, Git and virtual environments. Write clean, modular Python.
Then go deep on NumPy and Pandas:
- Vectorization and multi-indexing
- Advanced data cleaning strategies
- SQL with CTEs and window functions
- Pulling data from REST APIs
- Storing data fast with Parquet and Feather
Level 2: Applied Machine Learning & Deployment
Now build models that hold up in the real world:
- Statistical foundations and feature engineering
- Handling highly imbalanced datasets
- Regression models through to Random Forests
- Gradient boosting with XGBoost, LightGBM and CatBoost
- Interpreting models with SHAP values
- Deploying models as live web services with FastAPI or Flask
What you'll build: hands-on labs, including:
- The Data Cleaning Lab: repair a completely broken corporate sales report using RegEx and advanced Pandas mapping
- The Automated Ingestion Pipeline: build a live data tracker that syncs API data into a structured database
- The Loan Default Classifier: handle highly imbalanced data with SMOTE to predict financial risk
- The Hyperparameter Tournament: use Optuna and GridSearchCV to push a model's accuracy higher
Every lab gives you something real for your portfolio. You'll have code you wrote, data you cleaned and models you deployed.
By the end, you'll go from raw, messy data to a deployed, explainable ML model on your own.
Ready to stop watching and start building? Enrol now and write your first line of clean Python today.
- Team Yotta Academy
Who this course is for:
- Aspiring Data Scientists & Analysts looking for a structured, hands-on portfolio builder.
- Python Developers who want to pivot their career into Machine Learning and MLOps.
- Business & Financial Analysts looking to upgrade from Excel to high-performance automated data pipelines.