Machine Learning and Data Science with AWS- Hands On

Create an S3 bucket, configure region, public access, encryption, and object settings, then upload datasets and integrate with AWS services like Glue, Redshift, or Lambda.

Learn to create an AWS Glue crawler that scans S3 data sources, discovers schemas, and creates a table in the Glue Data Catalog; configure sources, IAM role, and scheduling.

Configure the crawler's output database by naming it and setting optional options, connect data sources, and set frequency and schedule runs, then edit or delete crawlers as needed.

Customize schema for a newly created Glue database by editing table columns, data types, and properties, and monitor crawlers and log groups in CloudWatch for auditing and troubleshooting.

Learn to run SQL queries in Amazon Athena, select datasets created by Glue, and export query results to an Amazon S3 bucket, while managing databases, tables, and joins.

Learn to create and save custom queries in AWS Athena, run conditional and advanced queries across tables, save and reuse queries, download results, and prepare data for visualization in QuickSight.

Get started with AWS QuickSight by signing in, configuring IAM, preparing data, modeling, and creating visualizations and dashboards from sources like S3 or Redshift in the cloud.

Import and upload datasets from S3, Oracle, or GitHub, then visualize in quick site. Use group and value fields to drive aggregations, creating pie, bar, map, and line charts.

Learn to create a tree map chart and customize charts by adjusting size, color by price, and recoloring categories, then export the results as a pdf for dashboards.

Learn data preparation and cleaning by editing datasets, creating calculated fields, applying filters, and customizing data types to build reliable visualizations and charts.

Learn to create a calculated field using ceil and concat to derive new columns from profit and product data, enabling tax calculations and string concatenation for dashboards.

Create, edit, and delete calculated fields to build profit per unit and price per unit using ceiling and division, integrating filters later for focused visualizations.

Learn to create and manage filters and excluded fields to refine datasets, apply include/exclude rules, and use date ranges with inclusive/exclusive options and calculated fields.

Create map charts for Italy using a prepared dataset, apply filters and conditional formatting to highlight profit by country, and align colors across map and bar charts in a dashboard.

Learn to create and customize pivot tables to summarize large data sets, using rows, columns, and values (country, estates, profit, tax from profit), with conditional formatting and KPIs.

Demonstrate conditional formatting and pivot tables to highlight top performing states with green color tones, creating a dashboard for data visualization and storytelling from US and Canada data, including Ontario.

Explore creating a word cloud to visualize categorical data by size and color, using group by, measures like order quantity or profit, and customizable layout and colors.

Learn to create a funnel chart that ranks categories by order quantity in descending order, showing the top 30 or 20 categories with configurable formatting, data labels, percentages, and colors.

Build a Flask-based web frontend for an NLP-powered app, enabling users to input text, translate into target languages, detect sentiment, and perform parts-of-speech analysis.

Learn to build a backend for a machine learning app with Flask, adding spam detection, sentiment analysis, part of speech tagging, named entity recognition, and translation.

Create a Python translation function and configure the AWS CLI access token. Translate English text to a user-selected language and return the result as JSON.

Demonstrates building a translation web app with AWS Translate to convert input text into a chosen language, using language codes and multi-language support.

Build a sentiment analysis app with AWS Comprehend to extract sentiment from text, classifying into positive, negative, neutral, or mixed, and apply it to product reviews and marketing insights.

Explore building a sentiment analysis web app using Amazon Comprehend to classify text into positive, neutral, and negative moods, and customize the model to map sentiment scores for reviews.

Explore part of speech tagging and its role in natural language processing by building a Python app that labels words and enables named entity recognition and sentiment analysis.

Learn to recognize named entities in text using a detect entities function, identifying person, location, organization, and time, and extract them into a dictionary for a web app.

Explore Amazon Deep Composer, a machine learning framework that creates music from scratch with generative AI and GANs, offering model training, uploads, and composition via keyboard or console.

Learn to create and edit melodies in the music studio with a deep composer, record or import tracks, and apply machine learning and generative ai to produce outputs.

Learn to use the GAN algorithm in machine learning to generate additional instruments in deep composer, comparing CNN techniques and transformers, with tracks created from gain parameters.

Learn to generate and stretch melodies with transformers by expanding a short input track up to 20 seconds, adjusting sampling, nucleus threshold, and creative risk, and choose instrument types.

Select MuseGAN or U-net, configure training data, set epochs and learning rate, and start training on the cloud. Learn generator and discriminator roles, dataset setup, and model monitoring.

Design the front end to upload images and generate keywords with AWS Rekognition, using a Python SDK to detect labels and objects and set a 40% confidence threshold for tagging.

Learn to build and test a Python script that uploads an image to the S3 bucket, generates keywords from objects with confidence in a JSON output, and extracts only labels.

Create a Python script using boto3 to upload images to an S3 bucket, expose a /upload post route, and generate keywords from the uploaded image.

Integrate S3 uploads with a keyword generator to build a real-world web app, importing scripts into a Flask API and displaying generated labels alongside the uploaded image.