Master AutoML - Google Cloud - Vertex AI

Explore Google Cloud AutoML's low-code approach to building and deploying machine learning models, illustrated by an insurance regression example and a simple dataset upload, target variable selection, and rapid deployment.

Explore Google Cloud Vertex AI for no-code prediction on images, videos, and documents. Experience single-label and multi-label classification, sentiment analysis, and entity extraction.

Identify independent variables like water and fertilizer that shape plant height as the dependent variable, and apply this framework to real-world cases such as insurance charges and loan decisions.

Explore regression and classification concepts with practical examples, including linear and logistic regression, scatterplot, and the interpretation of equations, slope, intercept, and prediction using multiple variables.

Explore how to measure accuracy in classification and regression models using metrics like accuracy, AUC, R-squared, and mean absolute error.

Explore multiple linear regression on an insurance dataset, with age, bmi, children, region, and smoker as independent variables predicting insurance charges using Vertex AI AutoML.

Build a classification model to predict customer churn using 18 independent variables and a binary target, upload the dataset, train, evaluate accuracy, and deploy for online predictions.

Discover natural language processing as the intersection of computer science, AI, and human language, and learn to analyze text for sentiment, speech-to-text, spam detection, and information extraction.

Explore core nlp concepts including segmentation and sentence segmentation, word tokenisation, stemming and lemmatization, stop words, dependency parsing, and parts of speech to enable text understanding.

NLP in Vertex demonstrates training a multi-label text classification model, deploying it to an endpoint, and testing predictions with input statements to identify the corresponding personality type.

Train an image classification model with AutoML on Vertex AI by creating labeled data, training, and deploying to an endpoint to predict car presence in images.