Become Certified Machine Learning Professional

Identify variables as temporary storage for data and practice with a Jupyter Notebook. Explore Python data types—integers, floats, booleans, strings, and complex numbers—through practical examples.

Explores basic Python operators—arithmetic, relational, and logical—demonstrating their use in simple expressions on two values in a Jupyter Notebook, and explains Python comments using #.

Learn python keywords and identifiers: reserved words you cannot use as variable or function names, and about 35 keywords guiding identifier formation, including case sensitivity and underscores.

Learn how to use single, double, and triple quotes for strings in a Jupyter notebook, including multi-line strings, printing, and zero-based indexing with counted spaces and end index adjustment.

Master common string operations in a hands-on Jupyter notebook: measure length, convert to lower and upper case, replace characters, and find a substring’s position.

Master decision making statements using if-else logic, including if, elif, and else blocks, and control flow with while and for loops demonstrated in a jupyter notebook.

Explore how functions are blocks of code that perform a single task, defined with def, called with arguments, and return values through examples like adding ten or checking even numbers.

Learn about Python tuples as an ordered, immutable collection that stores heterogeneous data, access elements by index, slice ranges, and use len, +, *, min, and max.

Explore lists: square brackets, mutability, heterogeneous elements, index access, and operations such as append, pop, reverse, sort, insert, concatenation, and repetition.

Discover how dictionaries store key-value pairs in curly braces, are unordered, and support adding, updating, merging with another dictionary, removing items with pop, and retrieving keys and values.

Explore Python sets as unordered collections of unique elements, shown with curly braces, and learn core operations like add, remove, update for multiple elements, and set union and intersection.

Explore essential Python libraries—numpy for mathematical operations, pandas for data manipulation and analysis, and matplotlib for static, interactive, and animated visualizations—highlighting import and prebuilt functions that speed code.

Learn to work with numpy arrays in Python, creating multi-dimensional arrays, initializing with zeros, and using arange for range and stepping, demonstrated in Jupyter notebooks.

Generate random numbers within bounds using numpy, control counts and shapes, then stack arrays with vstack, hstack, and column_stack; perform intersect1d and setdiff1d to find and remove common values.

Explore numpy basics by performing sums, arithmetic operations, and axis-based additions; compute mean, median, and standard deviation, and save or load arrays for machine learning workflows.

Learn how to use pandas to work with a one-dimensional, labeled series object, including creating series, changing its index, slicing, and element-wise addition, paving the way to data frames.

Learn how pandas data frames serve as two-dimensional data structures, build them from dictionaries, and use head, tail, shape, describe, drop, and basic statistics like mean, median, min, and max.

Learn to use Matplotlib for data visualization by creating plots, histograms, bar plots, scatter plots, and pie charts in Jupyter notebooks.

Master feature engineering to clean, organize, and handle missing values, including outlier detection and one-hot encoding, as you prepare real-world datasets for machine learning.

Use a histogram to understand the data's normal distribution, then detect and remove outliers by calculating the mean and three standard deviations.

Detect and remove outliers using the percentile method on a height column. Set max and min thresholds via quantiles, then filter the data to create a cleaned sports data dataset.

Learn to handle missing data in Pandas by filling with zeros or specific values, using forward and backward fill, interpolation, and dropping rows, then replace unusual values with mapping.

Apply linear regression with a single variable to model the relationship between mileage and price, fit the model in a Jupyter notebook, and predict car prices for given mileages.

Apply linear regression to multiple variables to predict prices, using mileage and age as independent variables in a Jupyter Notebook, fitting the model and predicting outcomes.

Explore the predictive function in machine learning using a linear regression formula y = coefficient × x + intercept, and see how mileage predicts value in a Jupyter notebook.

Split data into training and testing sets with the train_test_split method, then fit a linear regression model using mileage and other variables to predict sell price.

Explore logistic regression for binary classification and contrast with multiclass scenarios, applying the model to salary and retention data, training on split data, and evaluating accuracy.

Apply logistic regression to multiclass classification using the iris dataset with three species, selecting features, splitting data into train and test, fitting a model, and making predictions.

Save and load a linear regression model in a Jupyter notebook using pickle and joblib, demonstrating model persistence with dump and load and comparing efficiency.

Explore the decision tree classifier to predict outcomes by asking a sequence of questions, using label encoder to convert categorical data to digits and predict salaries above 100k.

Explore how to apply a support vector machine for classification, regression, and outlines detection, using Jupyter Notebook to load datasets, encode labels, split data, train, and predict.

Random forest is an ensemble method that builds many decision trees and uses majority voting to classify data, improving accuracy and handling missing values.

Learn how k-fold cross validation helps choose the right model by evaluating performance across multiple folds, improving over holdout methods, and comparing logistic regression, SVM, and random forest.

Explore probability and conditional probability, then build a Titanic survival predictor with Gaussian Naive Bayes, including mean imputation and label encoding of sex.

Learn to build a spam detector with a Naive Bayes pipeline in a Jupyter notebook, including label encoding, train-test split, CountVectorizer, and Multinomial Naive Bayes for text classification.

Learn how ensemble learning via bagging builds multiple small datasets, reduces overfitting and variance, and aggregates predictions by majority vote, with examples on diabetes prediction.

Apply L1 and L2 regularization to address overfitting in a logistic regression model using the iris dataset, with data encoding, train-test splits, and model evaluation in Jupyter.

Explore k-means clustering, an unsupervised learning algorithm, to identify clusters in income data and flag potential fraud transactions using inertia and the fit method.

Define deep learning as a subset of machine learning that mimics the human brain with neural networks of interconnected neurons, and begin using TensorFlow and Python to practice.

Install TensorFlow on your system using Anaconda and Python, ensuring only one Python installation. Open Git Bash as administrator and run 'pip install TensorFlow' to complete the setup.

Explore basic metrics operations, including subtracting expenses from revenue to compute profit and multiplying price per unit by units sold to derive revenue, with numpy in a Jupyter notebook.

Apply neural networks to customer churn prediction by preprocessing telco data with pandas, encoding features, scaling, and training a Keras sequential model; evaluate with precision, recall, and F1.

Explore precision, recall, and F1 score in a dog-detection dataset, and how true positives, false positives, true negatives, and false negatives shape accuracy.

Clean and organize real estate data for price prediction by dropping unused features, imputing missing values, engineering size and price per square foot, and removing outliers.

Apply the model to the data by encoding location with a label encoder and preparing X and Y for linear regression. Validate with cross-validation and grid search, then predict prices.

Learn to build an image classification model with OpenCV to recognize celebrities using a sample dataset, and explore data collection methods: manual download, web scraping, chrome extensions, and data vendors.

Detect and clean training data by filtering images to those with two visible eyes using OpenCV face and eye cascades, after setting up dependencies and uploading data.

Apply wavelet-based feature engineering to images by extracting features, combining color and wavelet representations, scaling images, and encoding labels as digits for model training.

train a classification model with a standard-scaler pipeline, perform grid search to compare svm, random forest, and logistic regression, and save the sport vector machine as the best performer.