Python Weekly (Issue 618 September 21 2023)

Python Weekly - Issue 618

September 21, 2023

Python Weekly

Welcome to issue 618 of Python Weekly. Let's get straight to the links this week.

From Our Sponsor 

A weekly newsletter featuring the best hand curated news, articles, tutorials, talks, tools and libraries etc for programmers.

Articles, Tutorials and Talks

ML models can be compiled to graphs, which can be traversed to perform forward and backward passes. This approach can improve performance and make it easier to debug ML models.

This article focuses on improving the modeling performance of LLMs by finetuning them using carefully curated datasets. Specifically, this article highlights strategies that involve modifying, utilizing, or manipulating the datasets for instruction-based finetuning rather than altering the model architecture or training algorithms (the latter will be topics of a future article). This article will also explain how you can prepare your own datasets to finetune open-source LLMs.

This post explores the proliferation of Python dataframes, dissecting the reasons behind their prevalence in data science and analysis, shedding light on the various libraries and frameworks that contribute to their abundance.

This  article delves into the creation of a handmade Transformer model, providing a detailed walkthrough of building this popular deep learning architecture from scratch, offering insights into its inner workings and structure.

Here are all the videos for the conference, brought to you by the EuroPython 2023 Team and the EuroPython Society.

This article discusses the integration of serverless functions with Django, highlighting how developers can leverage the benefits of serverless computing for specific tasks in a Django application. It explores the advantages of serverless architecture and provides practical insights for implementation.

Use Python to solve this classic probability puzzle that has stumped mathematicians and Nobel Prize winners!

This video discusses the problem of handling exceptions in Python when using Rust. In Rust, errors are handled differently, using response types, while in Python, exceptions are used. The video demonstrates how to raise exceptions in Python using Rust by creating a Python exception instance and raising it. It also shows how to handle exceptions by using mapping functions to convert Rust errors into Python exceptions. 

Django’s template engine has an underappreciated builtins option that selects libraries to preload in every template. Making a library a builtin avoids the need for an explicit {% load %} tag whenever you use its tags or filters. Putting key libraries in builtins can shorten your templates and make development a little bit faster. In this post, we’ll cover how to add a template library to builtins and remove existing {% load %} tags from your templates.

A worked out example: optimizing low-level code to get significant performance and memory improvements.

Learn Python programming in this complete course for beginners. This tutorial features mini-projects throughout so you can put what you learn into use immediately.

Lincoln Loop optimized a large publishing platform's database performance. Overall, the database performance increased 19 times.

In this guide, we will learn how to develop and productionize a retrieval augmented generation (RAG) based LLM application, with a focus on scale, evaluation and routing. 

Interesting Projects, Tools and Libraries

Generation of protein sequences and evolutionary alignments via discrete diffusion models.

Heart Rate Variability Training with the Polar H10 Monitor.

Galactic provides cleaning and curation tools for massive unstructured text datasets. It's designed to help you curate fine-tuning datasets, create document collections for retrieval-augmented generation (RAG), and even perform deduplication of web-scale datasets for LLM pre-training. This

Logparser provides a machine learning toolkit and benchmarks for automated log parsing, which is a crucial step for structured log analytics

MultiPlatform HTTP Reverse Shell.

The Security Toolkit for LLM Interactions.

Temporian is a Python library for feature engineering and data augmentation of temporal data (e.g. time-series, transactions) in machine learning applications.

Visual Pandas Selector: Visualize and interactively select time-series data.

QuasiQueue is a MultiProcessing library for Python that makes it super easy to have long running MultiProcess jobs. QuasiQueue handles process creation and cleanup, signal management, cross process communication, and all the other garbage that makes people hate dealing with multiprocessing.

Configurable Generation of Schemas and Knowledge Graphs at Your Fingertips.

Open source obd2 car diagnostics program.

Apple II emulator in Python.

A pythonic library providing light-weighted interface with LLMs

Python package for real estate scraping, supporting Zillow, Realtor.com & Redfin.

New Releases

Upcoming Events and Webinars

There will be following talks

  • Outside Technology: Building bridges between engineers and non-engineers

  • Rapid prototyping with Django

There will be following talks

  • The two language problem in vectorized python expressions

  • AI image generative algorithm in fashion design

There will be following talks

  • Machine-learning and causal inference: Toward new clinical evidence?

  • Project Feedback: Reduce your permissions management time while protecting your users!

There will be following talks

  • How we sped up our Django test-suite 10x

  • Python Generators: Light on Memory, Heavy on Power

  • Deadcode - a tool to find and fix dead (unused) Python code

There will be following talks

  • How to realize value from AI by bridging cross-domain knowledge to develop the right solutions?

  • Automate everything: using Python to extend the power of dbt at Voi

Our Other Newsletters

 - A free weekly newsletter for programmers.

- A free weekly newsletter for entrepreneurs featuring best curated content, must read articles, how to guides, tips and tricks, resources, events and more.