AI Music: It's WAY Better Than You Think - YouTube
Explore hundreds of AI Tools: https://futuretools.io/
Sunday, March 31, 2024
Pi: Personal AI; Mustafa Suleyman interview, book; Microsoft
Mustafa Suleyman: The AI Pioneer Reveals the Future in 'The Coming Wave' | Intelligence Squared - YouTube
Mustafa Suleyman is the ultimate AI insider. As co-founder of DeepMind and Inflection AI, he is one of the pioneers of the artificial intelligence revolution, potentially the single greatest accelerant of progress in history .Investors in Inflection AI include the likes of Microsoft, Nvidia and Bill Gates.
"over the last 10 years every single year the amount of compute
that we have used for The Cutting Edge AI models has grown by 10x so 10x 10x
10x 10x 10 times in a row now that is unprecedented in technology history
nowhere else have we seen a trajectory anything like that over the next five
years We'll add probably three or four orders of magnitude basically another thousand times the compute that you see
used today to produce gbt4 or the chat model that you might interact with..."
Inflection AI Introduces Pi, Your Personal AI
Amazon.com: The Coming Wave: Technology, Power, and the Twenty-first Century's Greatest Dilemma eBook : Suleyman, Mustafa: Kindle Store
Publication date : September 5, 2023
Publication date : September 5, 2023
The company was founded by entrepreneurs Reid Hoffman, Mustafa Suleyman and KarΓ©n Simonyan in 2022. The company has collaborated with Nvidia to develop hardware for generative artificial intelligence.[7]
In June 2023, the company raised $1.3 billion USD at $4 billion valuation.[7]
In March 2024, Suleyman and Simonyan announced their departure from the company in order to start Microsoft AI
performance: One Billion Rows Challenge
strange thing: using strings where binary format would be much more appropriate...
anyway, interesting code challenge
The One Billion Row Challenge (1BRC) is intended to be a fun exploration of how far modern Java can be pushed for aggregating one billion rows from a text file. .
What is the 1BRC challenge?
Input: A text file containing temperature values for a range of weather stations. Each row is one measurement in the format <string: station name>;<float: measurement>.Output: For each unique station, find the minimum, average and maximum temperature recorded and emit the final result on STDOUT in station name’s alphabetical order with the format {<station name>:<min>/<average>/<max>;<station name>:<min>/<average>/<max>}.
gunnarmorling/1brc: 1️⃣ππ️ The One Billion Row Challenge -- A fun exploration of how quickly 1B rows from a text file can be aggregated with Java : 1BRC on the Web
links to experiments on MANY languages and platforms!
Excellent learning resources.
Excellent learning resources.
But what happens if you do it in Go?
Saturday, March 30, 2024
OpenAPI codegen tools
openapi-typescript generates TypeScript types from static OpenAPI schemas quickly using only Node.js. It is fast, lightweight, (almost) dependency-free, and no Java/node-gyp/running OpenAPI servers necessary. (MIT license)
Tools for consuming OpenAPI schemas in TypeScript.
npm i -D openapi-typescript # supports Swagger/OpenAPI v3
# or
npm i -D openapi-typescript@5 # supports OpenAPI v2
npx openapi-typescript ./path/to/my/schema.yaml -o ./path/to/my/schema.d.ts
# π ./path/to/my/schema.yaml -> ./path/to/my/schema.d.ts [7ms]
npx openapi-typescript https://myapi.dev/api/v1/openapi.yaml -o ./path/to/my/schema.d.ts
npx openapi-typescript https://myapi.dev/api/v1/openapi.yaml -o ./path/to/my/schema.d.ts
# π https://myapi.dev/api/v1/openapi.yaml -> ./path/to/my/schema.d.ts [250ms]
it works, but the generated code is not quite "idiomatic" typescript.
it works, but the generated code is not quite "idiomatic" typescript.
alternatives
mapping local file swagger/index.yml to running docker container /tmp/index.yml
# unix bash
docker run -p 8000:8080 -e SWAGGER_JSON=/tmp/index.yml -v $(pwd)/swagger:/tmp --name=swagger-ui -d swaggerapi/swagger-ui
open http://localhost:8000
# windows powershell
docker run -p 8000:8080 -e SWAGGER_JSON=/tmp/index.yml -v ${pwd}/swagger:/tmp --name=swagger-ui -d swaggerapi/swagger-ui
start http://localhost:8000
# stop and remove
docker stop swagger-ui
docker rm swagger-ui
acacode/swagger-typescript-api: TypeScript API generator via Swagger scheme @GitHub
npx swagger-typescript-api -p ./swagger.json -o ./src -n myApi.tsGenerating typescript API client from Swagger
OpenAPI-Specification/3.0.3.md at main · OAI/OpenAPI-Specification
About Swagger Specification | Documentation | Swagger
example: Rustici Software SDK from Swagger codegen
mapping local file swagger/index.yml to running docker container /tmp/index.yml
# unix bash
docker run -p 8000:8080 -e SWAGGER_JSON=/tmp/index.yml -v $(pwd)/swagger:/tmp --name=swagger-ui -d swaggerapi/swagger-ui
open http://localhost:8000
# windows powershell
docker run -p 8000:8080 -e SWAGGER_JSON=/tmp/index.yml -v ${pwd}/swagger:/tmp --name=swagger-ui -d swaggerapi/swagger-ui
start http://localhost:8000
# stop and remove
docker stop swagger-ui
docker rm swagger-ui
EV: Tesla Cybertruck & Roadster reviews
Friday, March 29, 2024
save SVG (and HTML) as RDF graph
SVG is DOM, meaning it has a hierarchy of nodes, same as HTML, just as different XML variant.
Any any hierarchy is a "graph", and RDF designed for representing graphs.
So, technically SVG could be a "semantic web" visual graph.
Why? Just because it could be done, I guess :)
Old Dog, New Tricks by Kurt Cagle @ LinkedIn
Expressing SVG (and HTML) as RDF
There are many good use-cases for RDF variants, while SVG may not be one of them...
Visual graphs, expressed in higher-level model, that would be a very good use-case for RDF.
Resource Description Framework - Wikipedia
But RDF itself is already obscure enough, unless used in simplified serialization such as N3.
We may need to use AI to come up with a simpler to use graph representation syntax...
API specs: Swagger/OpenAPI, WADL, RAML
Microsoft Azure AI APIs have specs in Swagger (JSON), WADL (XML) and "English" format.
Cognitive Services APIs Reference
Cognitive Services APIs Reference
Another approach is to use WADL, which can be generated by some tools. In this case, it’s machine-readable, but definitely not human-readable. Also, writing a WADL manually is a tedious process.
Features | WADL | RAML | Swagger |
|---|---|---|---|
Release date | 2009 | 2013 | 2011 |
File format | XML | RAML | JSON/YAML |
Open source | Yes | Yes | Yes |
Commercial offering | No | Yes | Yes |
Language support | Java | JS, Java, Node, PHP, Python, and Ruby | Clojure, Go, JS, Java, .Net, Node, PHP, Python, Ruby, and Scala |
Authentication | No | Basic, Digest, OAuth 1, and OAuth 2 | Basic, API Key, and OAuth 2 |
API console | No | Yes | Yes |
Code generation for server (Java) | No | Yes | Yes |
Code generation for client (Java) | Yes | Yes | Yes |
Apache ECharts: JavaScript Visualization
Thursday, March 28, 2024
EV for $25K, now, worth it?
Fisker Ocean SUV: You can buy it for $25,000 right now. But there might be a catch | CNN Business
A brand new luxury electric SUV for the holy-grail price of $25,000. There just might be a slight catch.
The California-based electric SUV company Fisker faces a real possibility of bankruptcy,
The car itself has had some bad reviews. The Ocean was also the subject of a scathing review by American YouTube personality Marques Brownlee. His video was titled, “This is the Worst Car I’ve Ever Reviewed.”
This is the Worst Car I've Ever Reviewed - YouTube
A brand new luxury electric SUV for the holy-grail price of $25,000. There just might be a slight catch.
The California-based electric SUV company Fisker faces a real possibility of bankruptcy,
The car itself has had some bad reviews. The Ocean was also the subject of a scathing review by American YouTube personality Marques Brownlee. His video was titled, “This is the Worst Car I’ve Ever Reviewed.”
This is the Worst Car I've Ever Reviewed - YouTube
Node.js: The Documentary
Node.js: The Documentary | An origin story - YouTube
Back in 2008, most people thought of JavaScript as just a client-side language. But when Google's V8 appeared, young developer Ryan Dahl made the connection between non-blocking servers, V8, and JavaScript. It was by combining these key elements that he was able to create the now hugely popular Node.js. What young Ryan Dahl probably didn't expect was how much forking drama would follow.
Ryan Dahl (born 1981) is an American software engineer who is best known for creating the Node.js JavaScript runtime as well as the Deno JavaScript/TypeScript runtime
Back in 2008, most people thought of JavaScript as just a client-side language. But when Google's V8 appeared, young developer Ryan Dahl made the connection between non-blocking servers, V8, and JavaScript. It was by combining these key elements that he was able to create the now hugely popular Node.js. What young Ryan Dahl probably didn't expect was how much forking drama would follow.
Wednesday, March 27, 2024
Devin: AI software engineer? + OpenDevin; SWE-bench
Introducing Devin, the first AI software engineer
With our advances in long-term reasoning and planning, Devin can plan and execute complex engineering tasks requiring thousands of decisions. Devin can recall relevant context at every step, learn over time, and fix mistakes.Can Language Models Resolve Real-World GitHub Issues?
a sceptic view on this
- AI startup Cognition develops an AI software engineering platform that can automate entire projects.
- Devin has impressed big names in AI, like former Tesla AI director Andrej Karpathy.
- Well-funded by Peter Thiel and tech leaders, Cognition plans to offer early access to Devin for engineering teams soon.
AI + C programming mastery
a skeptical view on autonomy new programming tools...
and hopeful vision of future where such tools are helping productivity
AI just replaced us with Devin... seriously? Dr Chuck! - YouTube
a class mentioned that a good foundational knowledge of C programming
is a basis of mastery and ability to learn quickly any new tool or language
CC4E - C Programming for Everybody
this is a public version of essential C programming book
cc4e.com/book/
book & source code
csev/cc4e: C Programming for Everybody @GitHub
cc4e/book/code at master · csev/cc4e @GitHub
video "reading" or the book
Learn C Programming with Dr. Chuck (feat. classic book by Kernighan and Ritchie) - YouTube
previous popular course from the same author
PY4E - Python for Everybody
related course
C for Everyone: Programming Fundamentals Course (UCSC) | Coursera
Free online C (and other languages) compilers
- https://replit.com/languages/c
- PythonAnywhere - Use the Shell
- https://www.tutorialspoint.com/compile_c_online.php
- https://onecompiler.com/c
- Godbolt Compiler Explorer
C complier in Docker container image
- Docker-Hub-frolvlad/docker-alpine-gcc: The smallest Docker image with C compiler (GCC) (130MB)
- Dockerized build environments for C/C++ projects | Dmitry Danilov
- gcc - Official Image | Docker Hub
Original co-author of the referenced book
- The C Programming Language - Wikipedia
- Brian Kernighan - Wikipedia
- Dennis Ritchie - Wikipedia
- "C" Programming Language: Brian Kernighan - Computerphile - YouTube
- Ken Thompson interviewed by Brian Kernighan at VCF East 2019 - YouTube
- Ken Thompson - Wikipedia
some C language tutorials online
JavaScript speed: forEach, .map, .reduce vs for, for..of
Performance of JavaScript .forEach, .map and .reduce vs for and for..of
Both for and for..of are 3.5 times faster than reduce. However, the loops are much more verbose.
Both for and for..of are 3.5 times faster than reduce. However, the loops are much more verbose.
.forEach is almost the same as for or for..of, only 3 times slower.
Here we just calculate the sum of a and b for the whole array:
return array.reduce((p, x) => p + x.a + x.b, 0);Both for and for..of are 3.5 times faster than reduce. However, the loops are much more verbose:
let result = 0;for (const { a, b } of array) { result += a + b;}return result;Tuesday, March 26, 2024
cloud native: dapr: k8s demo
quickstarts/tutorials/hello-kubernetes at master · dapr/quickstarts @GitHub
This tutorial will get you up and running with Dapr in a Kubernetes cluster. You will be deploying the same applications from Hello World. To recap, the Python App generates messages and the Node app consumes and persists them. The following architecture diagram illustrates the components that make up this quickstart:
This tutorial will get you up and running with Dapr in a Kubernetes cluster. You will be deploying the same applications from Hello World. To recap, the Python App generates messages and the Node app consumes and persists them. The following architecture diagram illustrates the components that make up this quickstart:
Dapr - Distributed Application Runtime (dapr.io)
Monday, March 25, 2024
new JavaScript engines & runtimes
saghul/txiki.js: A tiny JavaScript runtime @GitHub (C)
txiki.js is a small and powerful JavaScript runtime. It targets state-of-the-art ECMAScript and implements many web platform features.
It's built on the shoulders of giants: it uses QuickJS-ng as its JavaScript engine and libuv as the platform layer
EV: $27K Skoda Epiq
New $25,000 Skoda Epiq is an Electric SUV that will destroy the Toyota Corolla - YouTube
why did they use name "epiq", already used by Hyundai
2023 Hyundai Ioniq 5 Epiq AWD review - Drive
this will be a year earlier than anticipated Tesla's model 2, for similar price
Redis -= Open Source license, alternatives: Redict, Garnet
Redis Adopts Dual Source-Available Licensing | Redis
all future versions of Redis will be released with source-available licenses. Starting with Redis 7.4, Redis will be dual-licensed under the Redis Source Available License (RSALv2) and Server Side Public License (SSPLv1). Consequently, Redis will no longer be distributed under the three-clause Berkeley Software Distribution (BSD).
redict/redict: A distributed key/value database - Codeberg.org @GitHub
Redict is a distributed key/value store (a remote dictionary, if you will).
Redict is an independent fork of Redis®* OSS 7.2.4 licensed under the Lesser GNU General Public license (LGPL-3.0-only). Redict is not affiliated with Redis®.
microsoft/garnet: Garnet is a remote cache-store from Microsoft Research that offers strong performance (throughput and latency), scalability, storage, recovery, cluster sharding, key migration, and replication features. Garnet can work with existing Redis clients.
links from
Changelog News #87 (2024-03-25)
all future versions of Redis will be released with source-available licenses. Starting with Redis 7.4, Redis will be dual-licensed under the Redis Source Available License (RSALv2) and Server Side Public License (SSPLv1). Consequently, Redis will no longer be distributed under the three-clause Berkeley Software Distribution (BSD).
redict/redict: A distributed key/value database - Codeberg.org @GitHub
Redict is a distributed key/value store (a remote dictionary, if you will).
Redict is an independent fork of Redis®* OSS 7.2.4 licensed under the Lesser GNU General Public license (LGPL-3.0-only). Redict is not affiliated with Redis®.
microsoft/garnet: Garnet is a remote cache-store from Microsoft Research that offers strong performance (throughput and latency), scalability, storage, recovery, cluster sharding, key migration, and replication features. Garnet can work with existing Redis clients.
- Garnet is a new remote cache-store from Microsoft Research, that offers several unique benefits:Garnet adopts the popular RESP wire protocol as a starting point, which makes it possible to use Garnet from unmodified Redis clients available in most programming languages of today, such as StackExchange.Redis in C#.
- Garnet offers much better throughput and scalability with many client connections and small batches, relative to comparable open-source cache-stores, leading to cost savings for large apps and services.
- Garnet demonstrates extremely low client latencies (often less than 300 microseconds at the 99.9th percentile) using commodity cloud (Azure) VMs with accelerated TCP enabled, which is critical to real-world scenarios.
- Based on the latest .NET technology, Garnet is cross-platform, extensible, and modern. It is designed to be easy to develop for and evolve, without sacrificing performance in the common case. We leveraged the rich library ecosystem of .NET for API breadth, with open opportunities for optimization. Thanks to our careful use of .NET, Garnet achieves state-of-the-art performance on both Linux and Windows.
links from
Changelog News #87 (2024-03-25)
Sunday, March 24, 2024
Microsoft Azure AI Solutions: AI-102
good Azure AI learning materials
Microsoft Certified: Azure AI Engineer Associate - Certifications | Microsoft Learn
Study guide for Exam AI-102: Designing and Implementing a Microsoft Azure AI Solution | Microsoft Learn
Course: AI-102 Microsoft Azure AI Solution Complete Exam Prep | Udemy Business
by Scott Duffy
+ related source
gottagetgit/AI102Files @GitHub, Python
gottagetgit/AI102Files-CSharp: Files for my AI-102 course @GitHub, C#
gottagetgit (gottagetgit) / Repositories
AI-102: Designing and Implementing a Microsoft Azure AI Solu | Udemy
by David Junior
Coursera
Exam Prep AI-102: Microsoft Azure AI Engineer Associate | Coursera
by "Whizlabs"
Microsoft Certified: Azure AI Engineer Associate - Certifications | Microsoft Learn
Study guide for Exam AI-102: Designing and Implementing a Microsoft Azure AI Solution | Microsoft Learn
- Azure AI services
- Azure AI Vision
- Azure AI Video Indexer
- Azure AI Language
- Azure AI Speech
- Azure AI Search
- Azure OpenAI
- Azure AI Document Intelligence
- Prepare to develop AI solutions on Azure
- Create and consume Azure AI services
- Secure Azure AI services
- Monitor Azure AI services
- Deploy Azure AI services in containers
- Analyze images
- Classify images
- Detect, analyze, and recognize faces
- Read Text in images and documents with the Azure AI Vision Service
- Analyze video
- Analyze text with Azure AI Language
- Build a question answering solution
- Build a conversational language understanding model
- Create a custom text classification solution
- Custom named entity recognition
- Translate text with Azure AI Translator service
- Create speech-enabled apps with Azure AI services
- Translate speech with the Azure AI Speech service
- Create an Azure AI Search solution
- Create a custom skill for Azure AI Search
- Create a knowledge store with Azure AI Search
- Plan an Azure AI Document Intelligence solution
- Use prebuilt Form Recognizer models
- Extract data from forms with Azure Document Intelligence
- Get started with Azure OpenAI Service
- Build natural language solutions with Azure OpenAI Service
- Apply prompt engineering with Azure OpenAI Service
- Generate code with Azure OpenAI Service
- Generate images with Azure OpenAI Service
- Implement Retrieval Augmented Generation (RAG) with Azure OpenAI Service
- Fundamentals of Responsible Generative AI
Course: AI-102 Microsoft Azure AI Solution Complete Exam Prep | Udemy Business
by Scott Duffy
+ related source
gottagetgit/AI102Files @GitHub, Python
gottagetgit/AI102Files-CSharp: Files for my AI-102 course @GitHub, C#
gottagetgit (gottagetgit) / Repositories
AI-102: Designing and Implementing a Microsoft Azure AI Solu | Udemy
by David Junior
Coursera
Exam Prep AI-102: Microsoft Azure AI Engineer Associate | Coursera
by "Whizlabs"
energy: Google's geothermal with fracking
Google’s Billion Dollar Gamble: You Won't Believe What it is! - YouTube
A New Type of Geothermal Power Plant Just Made the Internet a Little Greener | WIRED
Project Red is small—producing between 2 and 3 megawatts of power, or enough to power a few thousand homes—but it is a crucial demonstration of a new approach to geothermal power that could make it possible to harness the Earth’s natural heat anywhere in the world.
The Nevada site, an “enhanced” geothermal system, or EGS, works differently. Instead of drilling into a natural hydrothermal system, Fervo dug into rock that is completely dry and effectively created an artificial hot spring by pumping down water that returns to the surface much hotter.
That strategy piggybacks on hydraulic fracturing techniques developed by the oil and gas industry. Fervo drilled two wells that each extended more than 7,000 feet down before turning fully horizontal. I
A New Type of Geothermal Power Plant Just Made the Internet a Little Greener | WIRED
Project Red is small—producing between 2 and 3 megawatts of power, or enough to power a few thousand homes—but it is a crucial demonstration of a new approach to geothermal power that could make it possible to harness the Earth’s natural heat anywhere in the world.
The Nevada site, an “enhanced” geothermal system, or EGS, works differently. Instead of drilling into a natural hydrothermal system, Fervo dug into rock that is completely dry and effectively created an artificial hot spring by pumping down water that returns to the surface much hotter.
That strategy piggybacks on hydraulic fracturing techniques developed by the oil and gas industry. Fervo drilled two wells that each extended more than 7,000 feet down before turning fully horizontal. I
Saturday, March 23, 2024
Tempo.js vs moment.js, day.js
Tempo • Dates by FormKit
Tempo is a new library in a proud tradition of JavaScript date and time libraries.
Tempo is a new library in a proud tradition of JavaScript date and time libraries.
Inspired by the likes of moment.js, day.js, and date-fns,
Tempo is built from the ground up to be as small and easy to use as possible
— including first-class support for timezone operations.
Symbolic Math with Javascript: math.js
math.js | an extensive math library for JavaScript and Node.js
math.js has built-in support for symbolic computation (CAS). It can parse expressions into an expression tree and do algebraic operations like simplification and derivation on the tree.
math.js | an extensive math library for JavaScript and Node.js
math.js | an extensive math library for JavaScript and Node.js
mathjs - Libraries - cdnjs - The #1 free and open source CDN built to make life easier for developers
google/mathsteps: Step by step math solutions for everyone
A step by step solver for math
Nerdamer | Symbolic Math for Javascript
Nerdamer is a small and light-weight symbolic math expression evaluator written in javaScript. The parsing is done purely in javascript and uses no server-side program whatsoever. It can export to a pure javascript function, ideal for filtering user input.
Yaffle/Expression: A library to parse expressions, solve and simplify systems of linear equations, find eigenvalues and eigenvectors
It is a homemade library for JavaScript. It can parse expressions, solve and simplify systems of linear equations, find eigenvalues and eigenvectors, or calculate real roots of polynomials with integer coefficients for a specified accuracy.
Algebrite
Algebrite is a simple, comprehensible and extensible Javascript library for symbolic computation.
google/mathsteps: Step by step math solutions for everyone
A step by step solver for math
Nerdamer | Symbolic Math for Javascript
Nerdamer is a small and light-weight symbolic math expression evaluator written in javaScript. The parsing is done purely in javascript and uses no server-side program whatsoever. It can export to a pure javascript function, ideal for filtering user input.
Yaffle/Expression: A library to parse expressions, solve and simplify systems of linear equations, find eigenvalues and eigenvectors
It is a homemade library for JavaScript. It can parse expressions, solve and simplify systems of linear equations, find eigenvalues and eigenvectors, or calculate real roots of polynomials with integer coefficients for a specified accuracy.
Algebrite
Algebrite is a simple, comprehensible and extensible Javascript library for symbolic computation.
Friday, March 22, 2024
Azure AI learning paths
Browse all courses, learning paths, and modules - Training | Microsoft Learn
22 learning paths, a few estimated hours each, quite informative and well prepared materials
22 learning paths, a few estimated hours each, quite informative and well prepared materials
there is some duplication between similar paths
paths have same structure, with introduction, tutorials, a few questions, example and summary
example
Microsoft Certified: Azure AI Engineer Associate - Certifications | Microsoft Learn
paths have same structure, with introduction, tutorials, a few questions, example and summary
example
Microsoft Certified: Azure AI Engineer Associate - Certifications | Microsoft Learn
- Get started with Azure AI Services
- Develop decision support solutions with Azure AI Services
- Create computer vision solutions with Azure AI Vision
- Develop natural language processing solutions with Azure AI Services
- Implement knowledge mining with Azure Cognitive Search
Subscribe to:
Posts (Atom)