dotnet / ai-samples Goto Github PK
View Code? Open in Web Editor NEWLicense: MIT License
License: MIT License
Documentation Still Refers to 'Getting Started' folder, which has been Renamed
Today, the Python Evaluation building block can be used against a .NET backend that uses the Chat Protocol (Azure Search supports this). However, we know from customer feedback (A top request from MVPs) that they want a .NET native solution for this. The goal is to deliver .NET-specific building block documentation and sample code to support this. Currently this is still in a design phase, but the latest thinking is that we may adapt a community DotNet LLM Eval sample that was built by a fellow Microsoft employee (Maho Pacheco).
Work to be done:
• Evaluate options for LLM evaluation and land on the best practice guidance we’d give to devs. Decide if we’ll lead with the Python implementation for Build.
• Continue investigating what is needed for a .NET native evaluation block
• If continuing with adapting the DotNet LLM Eval sample, we’d need to:
o Likely provide an analysis UI that isn’t Grafana, for now aiming for parity with Pamela’s output.
o Figure out a solution for generating questions and answers to use as “grounded truth” – currently the Azure AI synthetics library, used by the Python block, does not have a .NET equivalent.
o Update to Support Semantic Kernel 1.x.
• Create guidance documentation
Should we add the model's response in as a ChatRequestAssistantMessage
? This way we can show it creating the chatbot-like experience?
We need to have a location that can support both gpt-35-turbo
and dall-e-3
we also need a gpt model that support tools call.
After looking at the docs https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models#gpt-35-turbo-model-availability
We will use
location: AusEast
gpt-35-turbo version: 1106
Create a full-stack app that showcases Aspire using components like:
Dependencies
Describe the issue:
Word spelling error.
Hike Images
samples (link1, link2) in the Semantic Kernel and Azure OpenAI SDK directories have this issue.
Suggestion:
Change to dall
.
Describe the issue:
Grammatical errors.
All samples in the Semantic Kernel and Azure OpenAI SDK directories have this issue.
Suggestion:
other samples
or another sample
.minute
.Great Idea!
Open GitHub discussion to Speed Up Iteration.
Privide links to Direct Users to Discords.
The link Getting-started: Deploying the Azure resources is broken
In samples that use models, provide similar samples that use hosted model providers:
The sample(s) should focus on only basic chat with a web UI, which can serve as a starting point template for future samples or new exploratory projects.
With this, we should provide a few different options for front-end UIs:
During the azd up
command you may encounter and error with like this:
'pwsh' is not recognized as an internal or external command
It's because no recent (v7+) PowerShell version has been installed on the Windows computer.
To fix the issue you need to execute the command dotnet tool update --global PowerShell
Let's update the documentation to prevent this error.
In the following document there is a reference to a project in that repository that doesn't exist.
ai-samples/src/llm-eval
https://learn.microsoft.com/en-us/dotnet/ai/tutorials/llm-eval @JakeRadMSFT @jmatthiesen
Local dev environment setup could be simplified using Winget configuration files.
Add Winget configuration to install the following prerequisites
https://github.com/dotnet/ai-samples/blob/main/src/quickstarts/README.md#prerequisites
Hi, I have this error while running the local-models/Phi, may I know how to resolve it?
My system is Ubuntu 22.04 and runs as CPU. I have tried to change to the latest version of TorchSharp 0.102.4, the error was the same.
Loading Phi2 from huggingface model weight folder
Unhandled exception. System.ArgumentOutOfRangeException: Length of JSON exceeded int.MaxValue, not supported yet (Parameter 'length')
at TorchSharp.PyBridge.Safetensors.LoadIndex(Stream stream)
at TorchSharp.PyBridge.PyBridgeModuleExtensions.load_safetensors(Module module, Stream stream, Boolean strict, IList1 skip, Dictionary
2 loadedParameters, Boolean leaveOpen)
at TorchSharp.PyBridge.PyBridgeModuleExtensions.load_safetensors(Module module, String location, Boolean strict, IList1 skip, Dictionary
2 loadedParameters)
at TorchSharp.PyBridge.PyBridgeModuleExtensions.load_checkpoint(Module module, String path, String checkpointName, Boolean strict, IList1 skip, Dictionary
2 loadedParameters)
at PhiForCasualLM.FromPretrained(String modelFolder, String configName, String checkPointName, ScalarType defaultDType, String device) in /home/xxxx/ai-samples/src/local-models/Phi/Phi.cs:line 45
at Program.
Thank you.
Work on CodeGen tokenizer needed for Phi-2 will be complete in Microsoft.ML.Tokenizers in the next few days.
Once it's available, consider replacing the current tokenizer implementation with the one provided by Microsoft.ML.Tokenizers.
https://github.com/dotnet/ai-samples/blob/main/src/local-models/Phi/Tokenizer.cs
We should help make sure folks create resources in the correct region and know how to choose model version.
Related PR Comment:
#26 (comment)
Model/Region Docs:
https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models#gpt-35-turbo-model-availability
Hi team,
LLamaSharp has already supported LLaMA2 and LLaMA3. Would you like to accept a PR which adds examples of LLaMA2 & LLaMA3 with it?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.