Certified Machine Learning Associate

Learn the fundamentals of artificial intelligence and machine learning, including supervised, unsupervised, and reinforcement learning, deep learning, neural networks, data collection, preprocessing, and AML applications.

Explore NumPy for arrays and calculations, Matplotlib for data visualization, and pandas for data manipulation and cleaning, including handling missing values and creating data frames.

Load and clean data, perform exploratory data analysis, handle missing values and outliers, apply feature scaling, normalization, encode categoricals, and split into training and testing sets, using iris data set.

Explore supervised learning with linear regression by loading a California housing data set, performing train-test split, fitting a linear model, and evaluating with mean squared error and R2 score.

Explain the difference between linear and logistic regression, train a logistic model on the iris dataset with a binary target, and plot the decision boundary for sepal length and width.

Learn to use decision trees and random forests on the iris data set, evaluate accuracy score and confusion matrix, apply cross-validation, and visualize trees with matplotlib.

Apply a support vector machine with an rbf kernel to classify iris data, using a standard scaler, then train, predict, evaluate with a confusion matrix and classification report.

Learn k-means clustering to form centroids and split data into k clusters using the iris data set in sklearn. Explore the elbow method, inertia, and centroid visualization.

Learn principal component analysis, a linear, unsupervised dimensionality reduction method that creates orthogonal components capturing the most data variance, enabling noise reduction, feature extraction, visualization, and faster, more accurate models.

Learn how an agent uses trial and error to maximize reward through reinforcement learning, applying states, actions, and a Q-learning approach on the frozen lake.

Explore deep learning foundations with neural networks, from perceptrons to backpropagation. Implement sigmoid activation and its derivative, train with gradient descent to adjust weights.

Explore deep learning with Keras and TensorFlow, building neural networks with layers, activations, and optimizers, applying feature extraction and classification to image, speech, and language tasks.

Explore convolutional neural networks (CNNs) that learn spatial hierarchies of features from images, using max pooling and dense layers for feature extraction and classification in deep learning.

Explore recurrent neural networks and their advanced LSTM variants for learning from sequential data, using numpy, TensorFlow, and Keras to forecast time series like sine waves.

Develop an ai-powered capstone project to detect plant diseases from leaf images and suggest fertilizer or pesticide by disease and region, using cnn, rnn, lstm, keras, and data preprocessing.