AI & Data Scientist – 100+ Lab Exercises (Basic, Intermediate, Advanced)

🔰 Basic Level (30+ Exercises)

 

Goal: Build a strong base in programming, statistics, and data handling.

Python Programming Essentials

  • Variables, loops, and conditional statements in Python.

  • Functions, classes, and object-oriented principles.

  • List/dictionary/set/tuple manipulations.

  • Reading/writing files and data parsing.

  • Exception handling and logging in Python.

Mathematics for Data Science

  • Linear algebra: vectors, matrices, dot/cross products.

  • Probability: distributions, Bayes theorem.

  • Statistics: mean, median, standard deviation, IQR.

  • Derivatives and gradients with visual plots.

  • Matrix decompositions: SVD, eigenvalues/eigenvectors.

Data Handling & Cleaning

  • Use Pandas for data preprocessing.

  • Handle nulls, outliers, duplicates.

  • Convert categorical features to numeric.

  • Merge, group, pivot, and filter dataframes.

  • Exploratory Data Analysis (EDA) on CSV/Excel data.

Visualization

  • Histograms, boxplots, scatterplots (Matplotlib/Seaborn).

  • Correlation matrix and pairplots.

  • Trend and line plots for time-series data.

  • Interactive charts using Plotly.

  • Geospatial mapping using Folium.

 

🚀 Intermediate Level (40+ Exercises)

 

Goal: Master core ML/AI concepts with practical applications.

Machine Learning Models (Supervised)

  • Implement Linear & Logistic Regression.

  • Train Decision Trees, Random Forests, XGBoost.

  • Model evaluation: confusion matrix, precision, recall, ROC.

  • Hyperparameter tuning using GridSearchCV.

  • Use scikit-learn pipelines for model building.

Unsupervised Learning

  • K-Means and DBSCAN clustering on real datasets.

  • PCA for dimensionality reduction and visualization.

  • Anomaly detection with Isolation Forest.

  • Market Basket Analysis with Apriori.

  • Hierarchical clustering and dendrogram analysis.

Natural Language Processing

  • Text cleaning, stemming, lemmatization.

  • Convert text to vectors using BoW, TF-IDF.

  • Build sentiment classifiers using Naive Bayes.

  • Topic modeling using LDA.

  • Named Entity Recognition using spaCy.

Deep Learning Fundamentals

  • Build simple neural networks using Keras/TensorFlow.

  • Activation functions: ReLU, sigmoid, softmax.

  • Train an image classifier on MNIST.

  • Use dropout and batch normalization.

  • Plot model accuracy/loss graphs.

 

🧠 Advanced Level (40+ Exercises)

 

Goal: Engineer intelligent solutions, deploy models, and leverage cutting-edge techniques.

Advanced AI Models

  • Implement CNNs for image recognition (e.g., CIFAR-10).

  • RNN and LSTM models for time-series and text data.

  • Build attention-based transformers from scratch.

  • Use BERT for text classification tasks.

  • Apply transfer learning using pre-trained models.

Generative AI & Foundation Models

  • Build a simple GAN to generate synthetic data.

  • Generate text using GPT and fine-tuned transformers.

  • Create image generation pipelines using Stable Diffusion APIs.

  • Build a multimodal AI model (text+image).

  • Prompt engineering for ChatGPT and Claude models.

Time-Series Analysis

  • Use ARIMA and SARIMA for forecasting.

  • Implement Prophet by Facebook for time-series trends.

  • Feature engineering for temporal features.

  • Anomaly detection in financial time-series.

  • Build dashboards with seasonal trends.

Big Data & Distributed AI

  • Data manipulation using PySpark.

  • Model training using MLlib on large datasets.

  • Parallelized model evaluation in Spark.

  • Build ML pipelines in Azure/AWS/GCP.

  • Ingest data from Kafka, store in Hadoop, analyze in Spark.

AI Deployment & MLOps

  • Build REST APIs for AI models using FastAPI.

  • Containerize models using Docker.

  • Implement model monitoring with Prometheus & Grafana.

  • Automate pipelines using CI/CD tools (GitHub Actions).

  • Setup MLFlow for experiment tracking.

Capstone Projects

  • AI-powered recommendation engine (e-commerce).

  • Fraud detection engine for financial transactions.

  • Medical diagnosis prediction system.

  • Generative chatbot using fine-tuned LLMs.

  • AI for predictive maintenance in manufacturing.

 

✅ Optional Tools & Platforms

 

  • Python, Jupyter, Colab, TensorFlow, PyTorch, Scikit-learn

  • Power BI, Tableau, Excel, BigQuery, Spark

  • HuggingFace Transformers, OpenAI APIs, MLFlow

  • Docker, Kubernetes, GitHub Actions, DVC

Download Syllabus 👇
Scroll to Top