{"data":{"_id":"6ae40d55ad69f3eadd1b3c07","name":"Data Science","slug":"data-science","image":"https://images.unsplash.com/photo-1551288049-bebda4e38f71?auto=format&fit=crop&w=1200&q=80","icon":"data-science","accentColor":"#0284C7","isFree":false,"price":1199,"features":["Modern NumPy & Polars Frameworks","Statistical Inference & Hypothesis Testing","Scikit-Learn, XGBoost & LightGBM","MLflow MLOps & Experiment Tracking"],"tags":["Interactive Roadmap","Data Science","Machine Learning","Polars","MLOps"],"category":null,"description":"Master modern data science from exploratory analysis with Polars and NumPy, to predictive modeling with XGBoost, model explainability (SHAP), and MLOps with MLflow.","modules":[{"icon":"🐻","title":"The Modern Data Science Stack: NumPy & Polars","description":"Vectorized computing, memory layouts, and Polars lazy evaluation query engines outperforming legacy Pandas.","duration":"2.5 hours","order":1,"isFree":true,"isPublished":true,"isProject":false,"content":"### High-Performance Dataframes\nPolars is written in Rust, leveraging Apache Arrow columnar memory and multi-threaded parallel execution for 10x-50x faster data wrangling.","subsections":[{"title":"Vectorized Computing with NumPy","description":"Broadcasting, multidimensional slicing, linear algebra operations, and memory views.","duration":"60 mins","content":"Core mathematical vectorization.","isFree":true,"isPublished":true,"order":1,"_id":"6ae3ed503f5d6ea8c6dd3c2b","locked":false},{"title":"Polars Lazy Evaluation & Query Optimization","description":"LazyFrame execution plans, predicate pushdown, and projection pushdown.","duration":"60 mins","content":"High-speed tabular data processing with Polars.","isFree":true,"isPublished":true,"order":2,"_id":"6ae3ed503f5d6ea8c6dd3c2c","locked":false}],"_id":"6ae3ed503f5d6ea8c6dd3c2a","locked":false},{"icon":"📊","title":"Exploratory Data Analysis & Statistical Inference","description":"Distributions, Central Limit Theorem, Hypothesis Testing (t-test, ANOVA, Chi-Square), and correlation analysis....","duration":"3 hours","order":2,"isFree":false,"isPublished":true,"isProject":false,"content":"","subsections":[{"title":"Probability Distributions & Central Limit Theorem","description":"Normal, Binomial, Poisson distributions and sample mean convergence.","duration":"75 mins","content":"","isFree":false,"isPublished":true,"order":1,"_id":"6ae3ed503f5d6ea8c6dd3c2e","locked":true},{"title":"Hypothesis Testing & Confidence Intervals","description":"Formulate null hypotheses, calculate p-values, and avoid Type I / Type II errors.","duration":"75 mins","content":"","isFree":false,"isPublished":true,"order":2,"_id":"6ae3ed503f5d6ea8c6dd3c2f","locked":true}],"_id":"6ae3ed503f5d6ea8c6dd3c2d","locked":true},{"icon":"⚙️","title":"Feature Engineering & Data Preprocessing Pipelines","description":"Imputation, Target / One-Hot Encoding, scaling, outlier mitigation, and Dimensionality Reduction (PCA)....","duration":"3 hours","order":3,"isFree":false,"isPublished":true,"isProject":false,"content":"","subsections":[{"title":"Preprocessing Pipelines in Scikit-Learn","description":"ColumnTransformer, custom transformers, and preventing data leakage between train/test splits.","duration":"75 mins","content":"","isFree":false,"isPublished":true,"order":1,"_id":"6ae3ed503f5d6ea8c6dd3c31","locked":true},{"title":"Dimensionality Reduction with PCA & t-SNE","description":"Eigenvalues, explained variance ratios, and high-dimensional clustering visualization.","duration":"75 mins","content":"","isFree":false,"isPublished":true,"order":2,"_id":"6ae3ed503f5d6ea8c6dd3c32","locked":true}],"_id":"6ae3ed503f5d6ea8c6dd3c30","locked":true},{"icon":"🌲","title":"Supervised Learning: Regressors, Classifiers & Trees","description":"Linear/Logistic regression with L1/L2 regularization, Decision Trees, Random Forests, and Gradient Boosting....","duration":"3.5 hours","order":4,"isFree":false,"isPublished":true,"isProject":false,"content":"","subsections":[{"title":"Regularized Linear Models & Logistic Regression","description":"Ridge (L2), Lasso (L1) for feature selection, and binary/multiclass log-loss.","duration":"75 mins","content":"","isFree":false,"isPublished":true,"order":1,"_id":"6ae3ed503f5d6ea8c6dd3c34","locked":true},{"title":"Gradient Boosting with XGBoost & LightGBM","description":"Histogram-based tree building, handling missing values natively, and early stopping.","duration":"90 mins","content":"","isFree":false,"isPublished":true,"order":2,"_id":"6ae3ed503f5d6ea8c6dd3c35","locked":true}],"_id":"6ae3ed503f5d6ea8c6dd3c33","locked":true},{"icon":"🔍","title":"Model Evaluation & Interpretability (SHAP & LIME)","description":"ROC-AUC, Precision-Recall curves, confusion matrices, SHAP values, and model explainability for stakeholders....","duration":"3 hours","order":5,"isFree":false,"isPublished":true,"isProject":false,"content":"","subsections":[{"title":"Evaluation Metrics for Imbalanced Datasets","description":"Why accuracy fails on 99:1 fraud datasets; using PR-AUC, F1-macro, and Cost Curves.","duration":"75 mins","content":"","isFree":false,"isPublished":true,"order":1,"_id":"6ae3ed503f5d6ea8c6dd3c37","locked":true},{"title":"Model Explainability with SHAP Values","description":"Game-theoretic feature attribution: global feature importance and local decision breakdowns.","duration":"75 mins","content":"","isFree":false,"isPublished":true,"order":2,"_id":"6ae3ed503f5d6ea8c6dd3c38","locked":true}],"_id":"6ae3ed503f5d6ea8c6dd3c36","locked":true},{"icon":"📦","title":"MLOps: Experiment Tracking with MLflow & Model Registry","description":"Log metrics, parameters, and artifacts with MLflow, version models, and build automated deployment pipelines....","duration":"3 hours","order":6,"isFree":false,"isPublished":true,"isProject":false,"content":"","subsections":[{"title":"MLflow Experiment Tracking & Artifact Storage","description":"Log hyperparameter runs, loss curves, and serialized model artifacts.","duration":"75 mins","content":"","isFree":false,"isPublished":true,"order":1,"_id":"6ae3ed503f5d6ea8c6dd3c3a","locked":true},{"title":"Model Registry & Deployment Workflows","description":"Staging to Production model transitions and serving models via FastAPI endpoints.","duration":"75 mins","content":"","isFree":false,"isPublished":true,"order":2,"_id":"6ae3ed503f5d6ea8c6dd3c3b","locked":true}],"_id":"6ae3ed503f5d6ea8c6dd3c39","locked":true},{"icon":"🏆","title":"Capstone: End-to-End Enterprise Churn & Customer Lifetime Value Pipeline","description":"Engineer a complete enterprise data science project predicting customer churn and lifetime value with Polars data wrangling, LightGBM modeling, SHAP e...","duration":"4.5 hours","order":7,"isFree":false,"isPublished":true,"isProject":true,"content":"","subsections":[{"title":"Data Wrangling, Feature Store & Model Training","description":"Engineer behavioral features and train cross-validated gradient boosted trees.","duration":"90 mins","content":"","isFree":false,"isPublished":true,"order":1,"_id":"6ae3ed503f5d6ea8c6dd3c3d","locked":true},{"title":"Model Explainability & Production API Serving","description":"Deploy the model behind a FastAPI prediction endpoint with SHAP explanations.","duration":"90 mins","content":"","isFree":false,"isPublished":true,"order":2,"_id":"6ae3ed503f5d6ea8c6dd3c3e","locked":true}],"_id":"6ae3ed503f5d6ea8c6dd3c3c","locked":true}],"details":[],"status":"published","__v":0},"message":"Course fetched successfully"}