Machine Learning Pro with Python
This advanced course provides a deep dive into machine learning using Python, covering essential topics such as linear algebra, statistics, exploratory data analysis, supervised and unsupervised learning, ensemble methods, NLP, time series forecasting, and an introduction to deep learning. Learners will gain hands-on experience with real-world data, popular Python libraries, and practical projects to build, evaluate, and deploy robust machine learning models for diverse applications.
Course Modules
Module 1: Linear Algebra with Python
- Numpy arrays in Python
- Matrix operations with python
- Eigen Values and Eigen Vectors using Python
- Dot product of vectors using python
- Distance metrics between two numerical vectors
- Similarity measures for numerical features; cosine similarity, correlation similarity
- Similarity measures for mixed features
Module 2: Statistics with Python
- Pandas features; Pandas dataframe, Pandas Series
- Descriptive statistics with python
- Data distributions using python
- Standardization of numerical data
- Normalization of numerical data
- Inferential statistics with python
Module 3: Exploratory Data Analysis-1
- Treatment of missing values
- Handling outliers
- Correlation
- Covariance
- Encoding of Categorical features; LabelEncoder, OneHotEncoder
Module 4: Unsupervised Machine Learning
- Clustering techniques; kMeans, Agglomerative, DBSCAN
- Dimensionality Reduction using PCA, t-SNE
- Market Basket Analysis; Apriori, fpgrowth, association rules
- Recommender Systems: Collaborative filtering, Content based filtering systems
Module 5: Supervised Machine learning-1
- Labelled data in Supervised Machine Learning; Classification and Regression
- Model Performance; Underfitting, Overfitting issues
- Model Validation Techniques; train-test-split, KFold CV, LOOCV, GridSearchCV
- Linear Regression using sklearn, statsmodels, regularization methods
- Logistic Regression using sklearn
- kNearestNeighbours Classification
- Support Vector Machines for Classification
- NaiveBayes Classifier
- DecisionTreeClassifier; Hyperparameters, pruning
Module 6: Supervised Machine learning-2: Ensemble Methods
- Bagging Principle; RandomForest
- Boosting Principle; GradientBoosting, xgboost
- Stacking Principle; StackingClassifier
Module 7: Exploratory Data Analysis-2
- Feature Selection; univariate methods, wrapper method, tree-based methods
- Anomaly detection using Isolation Forest
Module 8: Text Mining (NLP)
- Text Vectorization, WordCloud Visualization
- PoS tagging, Named Entity Recognition
- Sentiment Analysis
- Naïve Bayes Classifier for Text Classification
- Word Embedding Models
Module 9: Time Series Forecasting
- Time Series Concepts
- Time Series Visualizations
- Exponential Smoothing Methods
- Regression models; AR, MA, ARIMA, SARIMA
- Survival Analysis
Module 10: Introduction to Deep Learning
- ANN Architecture
- Activation functions
- Backpropagation and Gradient Descent algorithm
- Gradient Descent Variants
- Classification and Regression using ANN
- Vanishing and Exploding Gradients
- Regularization in ANN