From edeb98700a41621e370d9616fdf62cef56fafa70 Mon Sep 17 00:00:00 2001 From: LeeStott Date: Wed, 14 Aug 2024 16:59:06 +0100 Subject: [PATCH] Update --- README.md | 24 ++++++++++-------------- 1 file changed, 10 insertions(+), 14 deletions(-) diff --git a/README.md b/README.md index 42f38a7..65e63c2 100644 --- a/README.md +++ b/README.md @@ -35,12 +35,11 @@ Examples of successful model applications will provide insights into practical d We’ll wrap up with thoughts on where this field is headed and its potential impact. ## Technology Used -- LLMs, SLMs - [AI Studio](https://ai.azure.com) - [Azure Model Catalog](https://learn.microsoft.com/azure/ai-studio/how-to/model-catalog-overview) - [GitHub Model Catalog](https://github.com/marketplace/models) -- LLMs - GPT 3.5 GPT 4/4v/4o -- SLMs - Phi-3 ) +- Large Language Models - GPT 3.5 GPT 4/4v/4o +- Small Language Midelss - Phi-3 - [ONNXRuntime](https://onnxruntime.ai/) - [OLIVE](https://github.com/microsoft/OLive) - [Windows AI PC SDK](https://aka.ms/wcr) @@ -50,7 +49,7 @@ We’ll wrap up with thoughts on where this field is headed and its potential im ### Introduction (5 min) - Brief overview of Generative AI models - Importance of choosing the right model for specific tasks -- DEMO- Multimodal and GPT Prompts vs DALL-E Outcomes +- Multimodal and GPT Prompts vs DALL-E Outcomes ### Types of Generative AI Models (8min) - Large Language Models (LLMs) @@ -63,7 +62,7 @@ We’ll wrap up with thoughts on where this field is headed and its potential im Comparing SLMs vs LLMs Inference using text and vision building cross platform solution - [Notebooks](/src/01.InferencePhi3/01.notebooks/) -This demo takes an image png and then converts the image to code using Phi3 Onnx model local hosted vs GPT4o (Azure/GitHub Models Cloud hosted) the image is then converted to create a matplot python version of the image. +This demo takes an image png and then converts the image to code using Phi3 Onnx model local hosted vs GPT4o (Azure/GitHub Models Cloud hosted) the image is then converted to create a matplot python version of the image. - The opportunity of SLMs and LLMs @@ -80,7 +79,7 @@ This demo takes an image png and then converts the image to code using Phi3 Onnx - [**DEMO - Phi-3 Fine-tuning** (5 min)](/src/03.AIToolsSolutionE2E/Readme.md) -Cloud Based FineTuning using Azure AI Compute and Local based Fine Tuning using AI Toolkit +Cloud Based FineTuning using Azure AI Compute and Local based Fine Tuning using Microsoft Olive ### Tools for Model Evaluation and Comparison ( 5-8 min) - Azure Machine Learning for model accuracy measurement @@ -91,21 +90,19 @@ Cloud Based FineTuning using Azure AI Compute and Local based Fine Tuning using - Examples of successful model applications - Lessons learned from model deployment and usage -- [**DEMO - Phi-3 RAG using .NET Aspire** (5 min)](/src/04.CloudNativeRAG/Readme.md) +- [**DEMO - Cloud Native Distributed Application using Phi-3 & .NET Aspire to undertake RAG** (5 min)](/src/04.CloudNativeRAG/Readme.md) -RAG Aspire demo(We can deploy Phi-3 as Service and .using .NET Aspire to create Cloud Native Distribution Application) +RAG Aspire demo(Deployment of Phi-3 as Models as a Service and .using .NET Aspire to create Cloud Native Distribution Application) The RAG Aspire demo showcases the deployment of Phi-3 as a service and the use of .NET Aspire to create a cloud-native distributed application chat application. This demonstration aligns with Azure’s capabilities, highlighting the seamless integration and deployment of advanced AI models like Phi-3 within the Azure ecosystem. It also emphasizes the versatility of .NET Aspire in building scalable, cloud-native applications, catering to the growing demand for intelligent and responsive chat applications in various industries. - ### Conclusion (3 min) - Recap of key points - Final thoughts on the future of Generative AI models -### Q&A (5min) +### Q&A - Open floor for questions and discussion - ## Session Resources and Continued Learning | Resources | Links | Description | @@ -143,7 +140,7 @@ The RAG Aspire demo showcases the deployment of Phi-3 as a service and the use o | MMLU | MMLU ((Massive Multitask Language Understanding) evaluates how well the LLM can multitask | [https://github.com/hendrycks/test](https://github.com/hendrycks/test) | | KILT | Library for Knowledge intestive language tasks | [https://github.com/facebookresearch/KILT](https://github.com/facebookresearch/KILT) | -## Evaluation Frameworks +## Evaluation Frameworks | Frameworks / Platforms | Description | Tutorials/lessons | Reference | | -------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------- | @@ -189,5 +186,4 @@ The RAG Aspire demo showcases the deployment of Phi-3 as a service and the use o - - + \ No newline at end of file