Revised Syllabus: Python for Data Science & Deep Learning (10 Weeks)
Phase 1: Python Programming & Data Structures (Weeks 1–2)
Goal: Solid foundation in Python with emphasis on writing clean, efficient code for data tasks.
Week 1: Core Python & Scripting
- Variables, Data Types, Type Casting
- Control Structures: if, for, while, break, continue
- Functions: def, lambda, scopes, recursion
- Error Handling & File I/O
- List/Dict Comprehensions and Generators
Project: CLI To-Do App with File Storage
Week 2: OOP & Advanced Data Structures
- Classes, Objects, Encapsulation, Inheritance
- Data Structures: Lists, Tuples, Sets, Dicts
- JSON Handling, Datetime, Collections (deque, Counter)
Project: Address Book Manager with JSON Storage
Phase 2: Data Analysis with NumPy, Pandas, and Visualization (Weeks 3–4)
Goal: Master data wrangling, analysis, and visual storytelling.
Week 3: NumPy & Pandas Mastery
- NumPy arrays, indexing, broadcasting, random generation
- Pandas Series, DataFrames, slicing, merging, reshaping
- Data Cleaning: missing values, duplicates, type conversion
Project: EDA on Retail Sales Dataset
Week 4: Data Visualization (In-depth)
- Matplotlib: customization, subplots, themes
- Seaborn: histograms, KDEs, boxplots, heatmaps
- Plotly: interactive visuals, maps, animations
- Dash or Streamlit: dashboard creation
Project: Interactive Sales Dashboard with Plotly or Streamlit
Phase 3: Statistics, Probability & SQL (Weeks 5–6)
Goal: Gain a strong grip on the math behind data science and querying data.
Week 5: Applied Statistics & Probability
- Descriptive Statistics, Outlier Detection
- Distributions (Normal, Binomial, Poisson)
- Hypothesis Testing (t-test, chi-square)
- Correlation, Linear Regression, Confidence Intervals
Project: Medical Research A/B Testing Report
Week 6: SQL for Data Science
- SQL Basics, Joins, Filtering, Grouping
- CTEs, Subqueries, Window Functions
- SQLite with Python, SQLAlchemy
Project: Movie Database Analysis & Reporting with SQL
Phase 4: Machine Learning (Weeks 7–8)
Goal: Train and evaluate real ML models using Sklearn and deeper TensorFlow work.
Week 7: Supervised ML – Traditional Models
- ML Pipeline: preprocessing, train-test split
- Linear & Logistic Regression
- Decision Trees & Random Forests
- SVM, k-NN, Ensemble Methods
- Evaluation: Accuracy, ROC, AUC, F1, Precision/Recall
Project: Customer Churn Prediction + Model Comparison
Week 8: Deep Learning with Neural Networks
- Introduction to Neural Networks (Perceptron, Backpropagation)
- TensorFlow/Keras: Model, Layers, Compile, Train
- CNNs: Conv2D, Pooling, Dropout
- RNNs: LSTM/GRU for time-series
- Word Embeddings: Word2Vec, GloVe
Projects:
1. MNIST Digit Classification
2. Sentiment Classification using LSTM
3. Image Classifier with CNN (CIFAR-10)
Phase 5: Advanced Topics, APIs & Deployment (Weeks 9–10)
Goal: Productionize ML pipelines and use real-world data sources.
Week 9: APIs, Web Scraping & Big Data
- REST APIs with requests
- Web Scraping with BeautifulSoup & Selenium
- Dask for large-scale data processing
- Working with Spark DataFrames (Intro only)
Project: Real-time Crypto Dashboard with Live API Data
Week 10: ML Deployment & Capstone
- Flask/FastAPI for ML APIs
- Dockerize a Python App
- CI/CD: GitHub Actions, Unit Tests
- Streamlit Web Apps
- Portfolio Creation: GitHub, LinkedIn, Blogs
Capstone Project: End-to-End ML Project — Model + API + Dashboard
