Amazon Bedrock - Learn AI on AWS with Python!

Name: Amazon Bedrock - Learn AI on AWS with Python!
Rating: 4.5 (986 reviews)

Discover the power of Generative AI on AWS with Amazon Bedrock, create text and images with Python!

Created byJose Portilla, Pierian Training

Last updated 12/2023

English

English [Auto],Spanish [Auto],

What you'll learn

Discover the fundamentals of Amazon Bedrock's AI platform, gaining a comprehensive understanding of its architecture and capabilities.
Learn how to effectively install, configure, and utilize Amazon Bedrock's tools, including an in-depth exploration of text processing with Amazon Titan.
Master the use of advanced AI techniques such as Retrieval Augmented Generation (RAG), and how to apply embeddings and large language models for the real world.
Develop proficiency in extracting and processing complex information from diverse data sources, such as PDF documents and call transcripts
Understand the principles of AI-powered text and image processing, including the exploration of Stability AI parameters and Amazon's Boto3 for image generation
Apply your learned skills in practical, hands-on projects, including the creation of a visual recipe guide.

Course content

4 sections • 21 lectures • 3h 24m total length

COURSE DOWNLOADS AND FAQ0:39
Course Curriculum Overview3:28
Discover how to leverage Amazon Bedrock with Python to select and run text and image generation models, and implement retrieval augmented generation with vector databases and embeddings.
Installation and Set Up24:14
Install boto3 and configure AWS credentials by creating an IAM user and a credentials file. Test access by running a notebook to list S3 buckets.
Amazon Bedrock Overview13:54
Learn how Amazon Bedrock provides a managed gen ai platform with text, chat, and image models from Amazon, open source, and partners, including RAG and agents, via boto3 in Python.

Understanding Model Parameters7:51
Explore text generation parameters in Amazon Bedrock, including max token length, temperature, top P, and stop sequences, and learn how Titan and Llama 2 models differ in json configuration.
Amazon Titan17:29
Learn to connect to Amazon Titan via Bedrock runtime using boto3, and invoke Titan Text Express V1 with JSON input and generation options such as temperature, top_p, and max tokens.
Llama 212:29
learn to call the llama 2 model on amazon bedrock with boto3, noting prompts versus input text and model-specific parameter names. understand licensing thresholds above 700 million monthly active users.
Custom Models1:12
Code Along Project - Call Transcript10:46
Code along reads a call transcript, uses Bedrock runtime to summarize it in three sentences, and lists sales action items, while exploring prompt engineering and retrieval augmented generation concepts.
Ask Questions about PDF - Exercise4:03
Extract text from a pdf with python and pypdf2, then prompt an Amazon Bedrock model to answer the paid time off policy question: how many working days (25).
Ask Questions about PDF - Exercise Solution8:52
Learn to extract text from a corporate travel policy PDF using Pypdf2, clean newlines, and craft a Bedrock Titan prompt with context to answer user questions.

Understanding Image Generation Parameters4:48
Explore image generation on Amazon Bedrock with Titan and SD XL models, mastering prompts, CFG scale, seed, and image size; learn via the AWS console playground and model-specific nuances.
Stability AI Image Generation with Boto316:21
Explore Stability AI image generation on Amazon Bedrock using Boto3, crafting prompts with positive and negative weights, and tuning width, height, steps, seed, and CFG scale.
Titan Image Generation12:37
Explore Amazon's Titan image generator on Bedrock to transform text prompts into images, tune quality, height and width, and enable image variation with text to image and image variation features.
Titan Image Inpainting9:53
Explore image inpainting with the Titan image generator, using a source image and a mask to repaint selected areas, with prompts, negative prompts, and mask creation via OpenCV in Python.
Image Generation Exercise Overview2:41
Generate a dish and recipe from an ingredients list using a language model, then create an image and a watercolor variation with Titan image generation on Bedrock Runtime.
Image Generation Exercise - Solutions13:14
Connect to the bedrock runtime with boto3, craft prompts to turn ingredients into a dish and recipe, then generate a Quiche Lorraine image and apply a watercolor style.

Understanding RAG - Retrieval Augmented Generation8:58
Understand retrieval augmented generation using text embeddings and cosine similarity to retrieve documents and augment a language model prompt, offering cheaper results than fine-tuning and potential local deployment.
RAG - Example with Amazon Bedrock Embeddings and LLMs18:16
Discover how to implement retrieval augmented generation on Amazon Bedrock by embedding text, building a vector store, retrieving the most relevant context via cosine similarity, and prompting an LLM.
RAG Exercise Overview2:28
Walk through the Rag exercise notebook to build a Titan embedding function, read sports articles with pandas, compute cosine similarity, and form an LM with Rag to answer prompts.
RAG Exercise Solution9:50
Implement a rag workflow with Amazon Bedrock Titan embeddings, read and embed article texts with pandas, compute cosine similarity, and retrieve context for Titan Text Express v1.

Requirements

Python and Jupyter Notebook Experience Required
Credit Card Access for AWS (total spend of course is less than $1.00, but CC is required)

Description

Welcome to this course on Amazon Bedrock. This program has been expertly crafted to immerse you in the world of Amazon's AI platform, Bedrock, emphasizing practical Python applications. Suitable for both AI novices and seasoned practitioners, this course promises to deepen your understanding and provide hands-on experience in AI.

The course journey commences with an enlightening Section 1, offering an introduction to the course layout, essential resources, and FAQs. This foundational segment is essential for equipping you with the necessary tools and knowledge about Amazon Bedrock, including detailed installation and setup instructions to kickstart your AI adventure.

In Section 2, we delve into the complexities of Amazon Bedrock's text models. You'll explore critical text processing parameters and work with Amazon Titan and Llama 2, Bedrock's advanced text modeling tools. This section combines theoretical knowledge with practical application, featuring a project on call transcript analysis and exercises to enhance your skills in extracting and processing information from PDFs.

Section 3 transports you to the fascinating world of AI-powered image generation. It covers the essentials of image creation with Stability AI parameters and Amazon's Boto3, including Titan's capabilities in this area. The section's highlight is the Recipe Code Along Project, where you will creatively and technically generate a visual recipe guide.

The course culminates in Section 4, focusing on Retrieval Augmented Generation (RAG). This advanced topic is pivotal in AI, and you'll learn about its practical applications and benefits, particularly how Amazon Bedrock integrates embeddings and large language models in RAG.

Who this course is for:

Python developers interested in using Amazon Bedrock Models for Generative AI

Amazon Bedrock - Learn AI on AWS with Python!

What you'll learn

Explore related topics

Course content

Course Overview and FAQs4 lectures • 42min

Text Models with Amazon Bedrock7 lectures • 1hr 3min

Image Generation6 lectures • 1hr

RAG- Retrieval Augmented Generation4 lectures • 40min

Requirements

Description

Who this course is for: