TL; DR LangChainのメリデメを整理する過程で、今となってはopenai-pythonのうちChatGPTのAPIをを簡単に取り回せる程度のシンプルなライブラリがあるだけでも十分便利なんじゃないかと思ったので、ライブラリを個人で作ってみました。（バージョン0.0.1なのでちょっとお粗末な所もありますが） github.com はじめに こんにちは、データサイエンティストの坂元です。ABEJAアドベントカレンダーの13日目の記事です。世は大LLM時代ということで、ありがたいことにABEJAでも複数のLLMプロジェクトを推進させて頂いています。私自身もいくつかのLLMプロジェクトに参画しています。LLMといえばLangChainが便利ですね。OpenAI APIの利用だけでなく、各種ドキュメントのパースが出来たり、HuggingFaceやインデックスDBを扱う他のライブラリとインテ
Gamedev in Lisp. Part 1: ECS and Metalinguistic Abstraction - cl-fast-ecs by Andrew
Gamedev in Lisp. Part 1: ECS and Metalinguistic Abstraction In this series of tutorials, we will delve into creating simple 2D games in Common Lisp. The result of the first part will be a development environment setup and a basic simulation displaying a 2D scene with a large number of physical objects. It is assumed that the reader is familiar with some high-level programming language, has a gener
Over the course of my Spring 2020 semester at Harvey Mudd College, I developed a self-hosting compiler entirely from scratch. This article walks through many interesting parts of the project. It’s laid out so you can just read from beginning to end, but if you’re more interested in a particular topic, feel free to jump there. Or, take a look at the project on GitHub. Table of contents What the pro
If you combine this with the Unicode table, you’ll see that English is encoded with 1 byte, Cyrillic, Latin European languages, Hebrew and Arabic need 2, and Chinese, Japanese, Korean, other Asian languages, and Emoji need 3 or 4. A few important points here: First, UTF-8 is byte-compatible with ASCII. The code points 0..127, the former ASCII, are encoded with one byte, and it’s the same exact byt
python_modules.pdf
Python3 OpenCV / Pillow / pygame / Eel / PyDub / NumPy / matplotlib / SciPy / SymPy / gmpy2 / hashlib, passlib / Cython / Numba / ctypes / PyInstaller / curses / tqdm / JupyterLab / json / psutil / urllib / zenhan / jaconv Copyright © 2017-2025, Katsunori Nakamura 2025 8 19 Python ‘ .py’ Python Python Windows PSF Python py .py Enter macOS Linux PSF Python python3 .py Enter Anaconda Prompt Python p
はじめての自然言語処理 Fusion-In-Decoder でクイズに答えるモデルを作る | オブジェクトの広場
今回は Fusion-In-Decoder を使ってクイズに答えるモデルを作ります。以前から Wikipedia 等の外部情報を参照できるテキスト生成モデルを試してみたいと思っていました。Fusion-In-Decoder の発表は 2020 年なので少し前のモデルですが、T5 ベースで手軽に試せるサイズ感ですので、日本語で試してみましょう。 1. はじめに 今回紹介する Fusion-In-Decoder（以下、FiD ）1 は Meta AI (当時は Facebook AI Research) が発表した Open Domain question Answering タスクを解くテキスト生成モデルです。 じつは、以前から外部情報を参照できるテキスト生成モデルを試してみたくて2、 Google の RETRO3 の論文を読んでたんです。 なのですが、外部情報のサイズ感が 1000 B
Tier 4 Support § Support for these targets is entirely experimental. If this target is provided by LLVM, LLVM may have the target as an experimental target, which means that you need to use Zig-provided binaries for the target to be available, or build LLVM from source with special configure flags. zig targets will display the target if it is available. This target may be considered deprecated by
It is December. That means annoying Christmas things are everywhere, including but not limited to the annual programming semi-competition known as Advent of Code. The problem with Advent of Code is that it is a waste of time. Most of the puzzles are in the realm of either string processing (somewhat applicable to programming), logic puzzles (not really applicable to most programming), or stupid go
This paper reflects work done in late 2022 and 2023 to audit for vulnerabilities in terminal emulators, with a focus on open source software. The results of this work were 10 CVEs against terminal emulators that could result in Remote Code Execution (RCE), in addition various other bugs and hardening opportunities were found. The exact context and severity of these vulnerabilities varied, but some
Solving Quantitative Reasoning Problems With Language Models
Solving Quantitative Reasoning Problems with Language Models Aitor Lewkowycz∗, Anders Andreassen†, David Dohan†, Ethan Dyer†, Henryk Michalewski†, Vinay Ramasesh†, Ambrose Slone, Cem Anil, Imanol Schlag, Theo Gutman-Solo, Yuhuai Wu, Behnam Neyshabur∗, Guy Gur-Ari∗, and Vedant Misra∗ Google Research Abstract Language models have achieved remarkable performance on a wide range of tasks that require
A Lisp Interpreter Implemented in Conway’s Game of Life
Lisp in Life is a Lisp interpreter implemented in Conway’s Game of Life. The entire pattern is viewable on the browser here. To the best of my knowledge, this is the first time a high-level programming language was interpreted in Conway’s Game of Life. Running Lisp on the Game of Life Lisp is a language with a simple and elegant design, having an extensive ability to express sophisticated ideas as
Tier 4 Support § Support for these targets is entirely experimental. If this target is provided by LLVM, LLVM may have the target as an experimental target, which means that you need to use Zig-provided binaries for the target to be available, or build LLVM from source with special configure flags. zig targets will display the target if it is available. This target may be considered deprecated by
Join a VS Code Dev Days event near you to learn about AI-assisted development in VS Code. Release date: August 7, 2025 Update 1.103.1: The update adds GPT-5 prompt improvements, support for GPT-5 mini, and addresses these issues. Update 1.103.2: The update addresses these issues. Downloads: Windows: x64 Arm64 | Mac: Universal Intel silicon | Linux: deb rpm tarball Arm snap Welcome to the July 2025
I’m a fairly frequent Hacker News lurker, especially when I have some other important task that I’m avoiding. I normally head to the Active page (lots of comments, good for procrastination) and pick a nice long discussion thread to browse. So over time I’ve ended up with a good sense of what topics come up a lot. “The Bay Area is too expensive.” “There are too many JavaScript frameworks.” “Bootcam
GIMP 2.99.12 is a huge milestone towards GIMP 3.0. Many of the missing pieces are getting together, even though it is still a work in progress. As usual, issues are expected and in particular in this release which got important updates in major areas, such as canvas interaction code, scripts, but also theming… “CMYK space invasion”, by Jehan (based on GPLv3 code screencast), Creative Commons by-sa
Transformer models: an introduction and catalog — 2023 Edition January 16, 2023 52 minute read This post is now an ArXiV paper that you can print and cite. Update 05/2023 Another pretty large update after 4 months. I was invited to submit the article to a journal, so I decided to enlist some help from some LinkedIn colleages and completely revamp it. First off, we added a whole lot of new models,
For demonstration purposes, we'll fine-tune the multilingual version of the small checkpoint with 244M params (~= 1GB). As for our data, we'll train and evaluate our system on a low-resource language taken from the Common Voice dataset. We'll show that with as little as 8 hours of fine-tuning data, we can achieve strong performance in this language. 1{}^11 The name Whisper follows from the acronym
Try Mastering Emacs for free! Are you struggling with the basics? Have you mastered movement and editing yet? When you have read Mastering Emacs you will understand Emacs. It’s that time again: there’s a new major version of Emacs and, with it, a treasure trove of new features and changes. Notable features include the formal inclusion of native compilation, a technique that will greatly speed up y
vim9script4pythondevelopers.md Vim9 script for Python Developers Vim9 script�Vim script��������������������������������������������������系��� def������義����������Vim script��vim9script�����使����������(vim9script���
< Back Leaking the email of any YouTube user for $10,000 2025-02-12 Some time ago, I was looking for a research target in Google and was digging through the Internal People API (Staging) discovery document until I noticed something interesting: "BlockedTarget": { "id": "BlockedTarget", "description": "The target of a user-to-user block, used to specify creation/deletion of blocks.", "type": "objec
Introduction: Density Is King (With a Tiny VM) I've previously come to the conclusion that there's little reason for using bytecode in the modern world, except in order to get more compact code, for which it can be very effective. So, what kind of a bytecode engine will give you more compact code? Suppose I want a bytecode interpreter for a very small programming environment, specifically to minim
Amongst the many consequences of COVID-19 has been the suspension of in-person talks: suddenly, people like me have had to think about how to produce prerecorded videos. In this article I’m going to explain what I’ve come to understand about video recording and the “automatic video editing” technique I’ve used for videos such as Virtual Machine Warmup Blows Hot and Cold. To give you an idea of how
Plan 9 Desktop Guide
PLAN 9 DESKTOP GUIDE INDEX What is Plan 9? Limitations and Workarounds Connecting to Other Systems VNC RDP SSH 9P Other methods Porting Applications Emulating other Operating Systems Virtualizing other Operating Systems Basics Window Management Copy Pasting Essential Programs Manipulating Text in the Terminal Acme - The Do It All Application Multiple Workspaces Tiling Windows Plumbing System Admin
A Deep Dive into eBPF: Writing an Efficient DNS Monitoring.
eBPF / XDP is an in-kernel virtual machine, provides a high-level library, instruction set and an execution environment inside the Linux kernel. It’s used in many Linux kernel subsystems, most prominently networking, tracing, debugging and security. Including to modify the processing of packets in the kernel and also allows the programming of network devices such as SmartNICs. Use cases in eBPF im
Large Text Compression Benchmark Matt Mahoney Last update: July 3, 2025. history This competition ranks lossless data compression programs by the compressed size (including the size of the decompression program) of the first 109 bytes of the XML text dump of the English version of Wikipedia on Mar. 3, 2006. About the test data. The goal of this benchmark is not to find the best overall compressi
Amazon S3とAmazon CloudFrontによる静的ウェブサイトにSSL/TLS証明書(AWS Certificate Manager)・基本認証(Lambda@Edge)・IP制限(AWS WAF)をクロスリージョンで追加するAWS CloudFormationテンプレートとAWS Lambdaカスタムリソース - NRIネットコムBlog
小西秀和です。 この記事は過去に投稿した次の記事の続編で、SSL/TLS証明書(AWS Certificate Manager)、基本認証(Lambda@Edge)に加えてIP制限(AWS WAF)を追加したパターンでAmazon S3とAmazon CloudFrontによる静的ウェブサイトホスティングをAWS CloudFormationテンプレートとAWS Lambdaカスタムリソースを使用して構成するものです。 AWS LambdaカスタムリソースでSSL証明書・基本認証・CloudFrontオリジンフェイルオーバーを作成するAWS CloudFormationスタックを別リージョンにデプロイする 今回は更にOrigin Access Identity(OAI)をOrigin Access Control (OAC)に変更し、キャッシュポリシー(CachePolicy)、オリジンリ
LangChain Agentsを使ってテストコードから「テストの通るコード」を自動生成するプログラムを書いてみた｜mah_lab / 西見 公宏
先日以下のような記事を書いてみたものの、いちいち結果をChatGPTに手でコピペしながら検証するのはダサいなと思っていました。 そういうわけでRSpecが通るまで愚直に検証＆生成を繰り返すようなコードを書いてみたものの、修正履歴までChatGPTのコンテキストに持たせようとすると、すぐに最大トークン数を超えてしまい失敗してしまいます。 最大トークン数を超えないように頭の良いコンテキストを持たせるような実装も可能だとは思いますが、結構複雑な実装になってしまいそうです。 そんな中で出会ったのがこのツイートでした。 I saw a somewhat astonishing thing today. GPT was asked a question that it needed to write code to answer, and given access to a Python REPL.
Flipping Pages: An analysis of a new Linux vulnerability in nf_tables and hardened exploitation techniques
This blogpost is the next instalment of my series of hands-on no-boilerplate vulnerability research blogposts, intended for time-travellers in the future who want to do Linux kernel vulnerability research. Specifically, I hope beginners will learn from my VR workflow and the seasoned researchers will learn from my techniques. In this blogpost, I'm discussing a bug I found in nf_tables in the Linux
I find blockchain fascinating because it extends open source software development to open source + state. This seems to be a genuine/exciting innovation in computing paradigms; We don’t just get to share code, we get to share a running computer, and anyone anywhere can use it in an open and permissionless manner. The seeds of this revolution arguably began with Bitcoin, so I became curious to dril
LambdaLisp - A Lisp Interpreter That Runs on Lambda Calculus
LambdaLisp is a Lisp interpreter written as an untyped lambda calculus term. The input and output text is encoded into closed lambda terms using the Mogensen-Scott encoding, so the entire computation process solely consists of the beta-reduction of lambda calculus terms. When run on a lambda calculus interpreter that runs on the terminal, it presents a REPL where you can interactively define and e
Amazon Inspectorの検出結果をリソース単位でまとめてメール通知する方法 | DevelopersIO
はじめに 以前、Amazon Inspectorの検出結果をAWS Security Hubを経由してメール通知する方法をご紹介しました。 Inspectorは脆弱性を検出するたびに結果を作成するため、1つのリソースに対して複数の脆弱性が見つかった場合、その数だけメール通知が発生してしまいます。例えば、100個の脆弱性が検出されると、100通のメールが送信される状況でした。 本記事では、脆弱性ごとの通知ではなく、1リソースごとに検出結果をまとめて1回の通知で済ませる方法をご紹介します。 構成は以下のとおりです。 本実装は、以下の流れで処理を行います。 Inspectorで検出結果が作成され、Security Hub経由でEventBridgeが起動します。 Firehose ストリームに検出結果が一時保存されます。 Firehose ストリームは、指定したバッファ期間内に受信したイベントを
JupyterLab Changelog — JupyterLab 4.5.0a3 documentation
JupyterLab Changelog# v4.4# JupyterLab 4.4 includes a number of new features (described below), bug fixes, and enhancements. This release is compatible with extensions supporting JupyterLab 4.0. Extension authors are encouraged to consult the Extension Migration Guide which lists deprecations and changes to the public API. Code console improvements# The code console prompt can now be positioned on
Metashapeで全天球画像をキューブマップに変換
Metashapeは全天球画像(エクイレクタングラー)の処理に対応していますが、アライメント結果を他のSfM系ソフトで利用する場合は、全天球画像のままでは渡すことが出来ません。 postshot、RealityCapture、Inria 3D Gaussian Splatting等は全天球画像には対応していないためです。 しかし先日smert999氏が「Agisoft_metashape_convert_to_cubemap」を公開されました！ Metashapeでアライメントした全天球画像を、そのアライメント結果を維持したままキューブマップに変換できるというものです。 Theta, Insta360等で360度動画/全天球画像撮影 → Metashapeでアライメント → キューブマップ変換 → RealityCaptureで活用・postshotで3DGS生成 といったワークフローが可
Renato Athaydes
Revisiting Prechelt’s paper and follow-ups comparing Java, Lisp, C/C++ and scripting languages A discussion on programming languages' impact on productivity and program efficiency. In 1999, Lutz Prechelt published a seminal article on the COMMUNICATIONS OF THE ACM (October 1999/Vol. 42, No. 10) called Comparing Java vs. C/C++ Efficiency Differences to Interpersonal Differences, henceforth Java VS
A History of Clojure
71 A History of Clojure RICH HICKEY, Cognitect, Inc., USA Shepherd: Mira Mezini, Technische Universität Darmstadt, Germany Clojure was designed to be a general-purpose, practical functional language, suitable for use by professionals wherever its host language, e.g., Java, would be. Initially designed in 2005 and released in 2007, Clojure is a dialect of Lisp, but is not a direct descendant of any
Photo by Kai Dahms on UnsplashI love creating software libraries. Two months ago, I started porting one of our Python packages into a Rust crate. This new Rust crate matches the Python package’s ease of use and expressiveness. Along the way, I learned nine rules that can help you create beautiful libraries in Rust. The rules are: Create examples that don’t embarrass you.Accept all kinds of strings
GitHub - ComfyUI-Workflow/awesome-comfyui: A collection of awesome custom nodes for ComfyUI
ComfyUI-Gemini_Flash_2.0_Exp (⭐+172): A ComfyUI custom node that integrates Google's Gemini Flash 2.0 Experimental model, enabling multimodal analysis of text, images, video frames, and audio directly within ComfyUI workflows. ComfyUI-ACE_Plus (⭐+115): Custom nodes for various visual generation and editing tasks using ACE_Plus FFT Model. ComfyUI-Manager (⭐+113): ComfyUI-Manager itself is also a cu
Django for Startup Founders: A better software architecture for SaaS startups and consumer apps
In an ideal world, startups would be easy. We'd run our idea by some potential customers, build the product, and then immediately ride that sweet exponential growth curve off into early retirement. Of course it doesn't actually work like that. Not even a little. In real life, even startups that go on to become billion-dollar companies typically go through phases like: Having little or no growth fo
My birthday just passed, and to relax I wrote a parser combinator library. Over the last few years, I have worked quite a bit with Ningning Xie and Jeremy Yallop on parser combinators, which has led to a family of parser combinators which have optimal linear-time performance in theory, and which are many times faster than lex+yacc in practice. But these use advanced multistage programming techniqu
j次のブックマーク
k前のブックマーク
lあとで読む
eコメント一覧を開く
oページを開く
LangChainを使わない - ABEJA Tech Blog
Kalyn: a self-hosting compiler for x86-64
The Absolute Minimum Every Software Developer Must Know About Unicode in 2023 (Still No Excuses!) @ tonsky.me
0.8.0 Release Notes ⚡ The Zig Programming Language
Advent of Code on the Nintendo DS
"�[31m"?! ANSI Terminal security in 2023 and finding 10 CVEs
0.10.0 Release Notes ⚡ The Zig Programming Language
July 2025 (version 1.103)
Hacker News folk wisdom on visual programming
GIMP - Development version: GIMP 2.99.12 Released
Transformer models: an introduction and catalog — 2023 Edition
Fine-Tune Whisper For Multilingual ASR with 🤗 Transformers
What's New in Emacs 28.1?
Vim9 script for Python Developers · GitHub
Leaking the email of any YouTube user for $10,000
bytecode interpreters for tiny computers ⁑ Dercuano
Laurence Tratt: Automatic Video Editing
Large Text Compression Benchmark
A from-scratch tour of Bitcoin in Python
Nine Rules for Elegant Rust Library APIs
Linear-time parser combinators