Developers can now fine-tune GPT-3 on their own data, creating a custom version tailored to their application. Customizing makes GPT-3 reliable for a wider variety of use cases and makes running the model cheaper and faster.
You can use an existing dataset of virtually any shape and size, or incrementally add data based on user feedback. With fine-tuning, one API customer was able to increase correct outputs from 83% to 95%. By adding new data from their product each week, another reduced error rates by 50%.
Can I create a syllabus with Node-RED?
Node-RED is a programming tool for wiring together hardware devices, APIs and online services in new and interesting ways. It provides a browser-based editor that makes it easy to wire together flows using the wide range of nodes in the palette that can be deployed to its runtime in a single-click.Node-RED
Here’s an idea: use the low code browser based flow design tools of Node-RED to create a syllabus of online resources. Since Node-RED can work with APIs and I have an API for CALI resources, I should be able to use Node-RED tools to design a syllabus as a flowchart through a series of resources.
I suspect there are a few missing pieces here like a program or routine to step through the flow created by Node-RED and a system from tracking results that talks to Node-RED. The pieces are likely buildable one identified. The result would be a programmed learning environment using Node-RED as a primary component.
Spot API is an issue spotter making use of NSMIv2 and built upon Learned Hands data
Spot is an NSMIv2 entity/issue spotter built by Suffolk University’s Legal Innovation and Technology (LIT) Lab. Spot builds upon data from the Learned Hands online game, a partnership between the LIT Lab and Stanford’s Legal Design Lab. Learned Hands aims to crowdsource the labeling of laypeople’s legal questions for the training of machine learning (ML) classifiers/issue spotters.
Source: Spot API Performance
In a nut shell you send a legal question and it returns a set of NSMIv2 tags. One could then use the those tags to link to resources that are using NSMI to tag/classify data. One catch here is that it is not clear how many public websites are using NSMI to tag content.
Stoplight Studio | OpenAPI Design, Planning & Modeling Tool
Stoplight Studio | OpenAPI Design, Planning & Modeling Tool https://stoplight.io/studio/
GraphQL | A query language for your API
GraphQL is a query language for APIs and a runtime for fulfilling those queries with your existing data. GraphQL provides a complete and understandable description of the data in your API, gives clients the power to ask for exactly what they need and nothing more, makes it easier to evolve APIs over time, and enables powerful developer tools.
Tutorial: How to build a GraphQL server – Apollo GraphQL
[H]ow to build a GraphQL server that connects to multiple backends: A SQL database, a MongoDB database and a REST endpoint. We’ll be combining all of them to build a very basic blog with authors, posts and views.
Combining Posts From Multiple WordPress Sites Using The REST API
Source: Combining Posts From Multiple WordPress Sites Using The REST API
Randall Degges – Why I Love Basic Auth
Let’s talk about Basic Auth:
- It’s a well and clearly defined specification.
- It’s been around since ~1996.
- It’s super simple.
Here’s the short version of how it works.
- You are a developer.
- You have an API key pair: an API Key ID and an API Key Secret. Each of these is a randomly generated string (usually a uuid).
- To authenticate against an API service, all you need to do is put your credentials into the HTTP Authorization header.
Amazon API Gateway – Build and Run Scalable Application Backends | AWS Official Blog
Amazon API Gateway – Build and Run Scalable Application Backends | AWS Official Blog https://aws.amazon.com/blogs/aws/amazon-api-gateway-build-and-run-scalable-application-backends/
Seldon is an open source recommendation platform
Seldon is made up of many components that work together to deliver the best recommendations. Roughly, all user actions are captured via the REST API and streamed to logs. Those logs are processed in batch and new user models are delivered to the API Server. Then recommendations are delivered via the REST API.
The code for Seldom is on Github at https://github.com/SeldonIO/seldon-server.