# dataframe (blogs) — RSS Amplifier Recent posts from the 6 feeds in the RSS Amplifier directory that cover dataframe. Page: Feed: --- ## [Not everything is AI or Agents](https://ogirardot.writizzy.blog/p/not-everything-is-ai-or-agents) _2026-05-06 · Olivier Girardot's Ramblings_ The boring stuff is the point ## [A first look at token efficiency](mchav.github.io/a-first-look-at-token-efficiency/) _2026-04-24 · Michael Chavinda_ A while ago I saw the article Which languages are most token efficient. The article was largely discredited since it didn’t have a clear methodology. Unfortunately, I haven’t come up with one yet but I thought to do a small, anecdotal experiment comparing code generation in Haskell notebooks (with Sabela) vs Python notebooks (Marimo). Notebooks are a great environment to use with agents since they… ## [Pandas feels clunky coming from R. What about Haskell?](mchav.github.io/being-less-clunky/) _2026-04-18 · Michael Chavinda_ Some years ago I came across an issue in the Frames repo that mentioned a blog post titled “Why pandas feels clunky when coming from R.” The article showed a side-by-side of simple data exploration in R and compared the code to Pandas. At the time, the author concluded that Pandas was “clunkier” than R. The author operationalises the definition of clunkiness but I think it’s really more of a… ## [Grow and mow: interpretable models with boosting, symbolic regression and e-graphs](mchav.github.io/grow-and-mow/) _2026-04-09 · Michael Chavinda_ This post is the convergence of two ideas that have been floating in my head for about a year. Can we learn messy stochastic models and use algorithmic/algebraic tools to rein in model complexity to make models interpretable? ## [Join dataframes in Scheme](https://www.travishinkelman.com/posts/join-dataframes-in-scheme.html) _2026-04-05 · Travis Hinkelman_ This post is part of a series on the dataframe library for Scheme (R6RS). In this post, I will contrast functions for joining dataframes from the dataframe library with the dplyr package for R. Set up First, let’s create a couple of dataframes in both languages. library (dplyr) df1 \<- data.frame ( name = c ( "Alice" , "Bob" , "Carol" ), age = c ( 30 , 25 , 35 )) df2 \<- data.frame ( name = c (… ## [Type-level programming is still programming](mchav.github.io/type-level-programming-is-still-programming/) _2026-04-04 · Michael Chavinda_ I was showing a friend the typed dataframe API. The whole pitch was: look, you derive a schema from your data, and then the compiler catches column name typos, type mismatches, all the stuff that would otherwise blow up at runtime. I had a nice demo ready using the Kaggle credit card fraud dataset (about 284,000 rows, 31 columns). ## [Display and glimpse dataframes in Scheme](https://www.travishinkelman.com/posts/display-and-glimpse-dataframes-scheme.html) _2026-04-01 · Travis Hinkelman_ This post is part of a series on the dataframe library for Scheme (R6RS). In this post, I will describe dataframe-display and dataframe-glimpse , which are inspired by the default print format for tibbles in R and the dplyr::glimpse function. Both procedures rely heavily on Chez Scheme’s format procedure, so I will first give an overview of the format directives used in the dataframe library… ## [What Category Theory Teaches Us About DataFrames](mchav.github.io/what-category-theory-teaches-us-about-dataframes/) _2026-03-28 · Michael Chavinda_ Every dataframe library ships with hundreds of operations. pandas alone has over 200 methods on a DataFrame. Is pivot different from melt? Is apply different from map? What about transform, agg, applymap, pipe? Some of these seem like the same operation wearing different hats. Others seem genuinely distinct. Without a framework for telling them apart, you end up memorizing APIs instead of… ## [Reverse engineering now and then](https://ogirardot.writizzy.blog/p/reverse-engineering-now-and-then) _2026-03-11 · Olivier Girardot's Ramblings_ Reverse engineering, hacking and cracking has a long tradition of being tedious and intensive - how has it changed with our new AI tools ## [Good software knows when to stop](https://ogirardot.writizzy.blog/p/good-software-knows-when-to-stop) _2026-03-05 · Olivier Girardot's Ramblings_ Good software knows what problem it solves and what needs to be tackled by another tool ## [Sync bookmarks across every browser (Sponsored)](https://crawlproof.com/a/wBPpNeON57MQ) _2026-03-04 · **Sponsored**_ Two-way sync across Chrome, Firefox and Safari; built-in ad and tracker blocker. ## [For a new golden age of FOSS](https://ogirardot.writizzy.blog/p/for-a-new-golden-age-of-foss) _2026-02-23 · Olivier Girardot's Ramblings_ Arguing that the current Generative AI trend is a chance for disrupting existing staled eco-systems with free software for the common good ## [Building An Ads Ranking System From Scratch](https://hongtaoh.com/en/2026/02/20/ads-ranking/) _2026-02-20 · Hongtao Hao_ We are building an Ads ranking ML system from scratch. The basic idea is this: think about Facebook or Instagram. The platform has millions of ads to show, but it needs algorithms to decide which ads to show. Why? Because, if it shows you ads randomly, you’ll get annoyed and leave the platform, let alone clicking the ads and purchasing. What Meat wants: You keep happy and like the ads, and… ## [Learning better decision tree splits - LLMs as Heuristics for Program Synthesis](mchav.github.io/learning-better-decision-tree-splits/) _2026-01-16 · Michael Chavinda_ A lot of tabular modeling gets easier the moment you stumble onto the right derived quantity. Not something mysterious or “deep.” It’s usually something you can name: a ratio that turns two raw columns into a rate; a difference that becomes a margin; a simple count that captures what a bunch of messy fields were hinting at. ## [CHTC GPU Lab: LLM Inference Example](https://hongtaoh.com/en/2026/01/14/chtc-gpu/) _2026-01-14 · Hongtao Hao_ A minimal example project for running Large Language Model (LLM) inference on CHTC GPU Lab at UW-Madison. Why This Project? CHTC’s documentation is great, but setting up LLM inference involves several gotchas: GPU Lab nodes cannot directly access /staging via paths in submit files Version conflicts between container PyTorch and pip-installed packages Efficient environment management to avoid… ## [Installing docker on a Chromebook](mchav.github.io/installing-docker-on-chromebook/) _2025-11-20 · Michael Chavinda_ I couldn’t find any instructions online so I thought I’d post them here for anyone who goes through a similar struggle. ## [An introduction to program synthesis (Part II) - Automatically generating features for machine learning](mchav.github.io/an-introduction-to-program-synthesis-part-ii/) _2025-11-04 · Michael Chavinda_ Introduction This post kicks off the second part of a hands-on series about program synthesis. We’ll apply the previously explored technique (an enumerative bottom-up search) to a slightly more realistic problem: automatically generating features for the Iris dataset. ## [Electric vehicles in California](https://www.travishinkelman.com/posts/electric-vehicles-california.html) _2025-10-23 · Travis Hinkelman_ I bought my first electric vehicle (EV) about a month ago and I thought it would be useful to capture my decision making around this purchase and initial impressions of driving an EV. I also thought it would be fun to use my Scheme dataframe library to explore data on EVs in California. The decision to buy an EV was relatively abrupt. I wanted to replace the 2012 Honda Civic that my son was… ## [Progress towards Kaggle-style workflows in Haskell](mchav.github.io/iris-classification-in-haskell/) _2025-10-13 · Michael Chavinda_ There’s been a lot of work in the Haskell ecosystem that has made it easier to write interactive Kaggle-like scripts. I’d like to showcase the synergy between 3 such tools: dataframe (my own creation), hasktorch, and IHaskell. ## [An introduction to program synthesis](mchav.github.io/an-introduction-to-program-synthesis/) _2025-09-11 · Michael Chavinda_ Introduction This post kicks off a hands-on series about program synthesis—the art of teaching machines how to generate code. We’ll build a tiny, FlashFill-style synthesiser that learns to turn strings like “Joshua Nkomo” into “J. Nkomo” from input/output pairs. We’ll see how to define a tiny string-manipulation language, write an interpreter, and search the space of programs to find one that… ## [My experiences at the Midwest ML Symposium (MMLS) 2025](https://hongtaoh.com/en/2025/06/25/mmls/) _2025-06-25 · Hongtao Hao_ I knew about this event only three days before it occurred. Luckily, Maria, one of the organizers, emailed me that I could still register. I did and it only cost $25 for students. On the first day of this event, I got up at around 4 AM and drove from Madison to Chicago. After I arrived, I parked at a nearby place and walked to the Logan center for the arts, where the event was held. ## [AI News Summaries & Audio (Sponsored)](https://crawlproof.com/a/sd3RqV11fTYX) _2025-06-25 · **Sponsored**_ Tap an article for an AI summary and an audio briefing. ## [My experiences at the Morgridge Entrepreneurial Bootcamp (MEB)](https://hongtaoh.com/en/2025/06/25/meb/) _2025-06-25 · Hongtao Hao_ I really enjoyed my experiences at MEB 2025 . It was very memorable, meaningful and helpful for me. MEB Program Overview The first day I was so tired and constantly sleepy during lectures. That was because of three reasons: 1) I didn’t sleep well the day before and 2) The lectures on the first day were boring to me; 3) Being among a large group of people was exhausting to me. ## [NetLogo: a retrospective](https://www.travishinkelman.com/posts/netlogo-retrospective.html) _2025-05-18 · Travis Hinkelman_ In this post, I briefly reflect on my experiences with NetLogo , which played a key role in my professional life. My meandering graduate school research experiences were all broadly related to predator-prey interactions, foraging behavior, and animal movement. During my PhD (the 2nd try) , I was studying the movement behavior of ladybug larvae after they consumed low-quality, potentially toxic,… ## [Updating Shiny Scorekeeper app with Shiny modules and bslib](https://www.travishinkelman.com/posts/shiny-scorekeeper-update.html) _2025-04-27 · Travis Hinkelman_ Shiny Scorekeeper is a basketball scorekeeper app built with the Shiny web framework for R . The app was initially built in 2018, but I recently decided to update it to improve maintainability and provide a more modern look. It’s sort of a strange choice to invest time in this project in 2025. I built the app for scoring videos of my son’s youth basketball games, but my son is no longer a youth… ## [How to Delete Large Files on Shared Linux Clusters](https://hongtaoh.com/en/2025/04/21/linux-large-delete/) _2025-04-21 · Hongtao Hao_ First, find the top largest files find ~ -type f -exec du -h {} + | sort -rh | head -n 20 Clean conda: conda clean -a -y , this will remove Tarballs and extracted packages, Index caches, and Logs Conda envs: List all envs: conda info --envs Remove one: conda env remove --name YOURENVNAME clean pip cache: pip cache purge ## [Understanding Kernel Density Estimation](https://hongtaoh.com/en/2025/04/17/kde/) _2025-04-17 · Hongtao Hao_ Matthew Conlen explains KDE well with a dynamic visualization , worth referring to. Origin KDE is a very important concept. It’s particularly useful when the true underlying distribution of data is uncertain and likely doesn’t follow a standard distribution (like a normal distribution). Let’s use the data from Gaussian Mixture Distribution as an example: import numpy as np import… ## [HT Condor Tricks](https://hongtaoh.com/en/2025/04/02/htcondor/) _2025-04-03 · Hongtao Hao_ I have been using HT Condor (through UW Madison’s CHTC) a lot. They are really amazing resources. Conda I just realized we can install conda for python enviroments management. # Download Miniconda installer wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86\_64.sh # Install Miniconda in your home directory bash Miniconda3-latest-Linux-x86\_64.sh -b -p $HOME/miniconda #… ## [Object oriented programming deemed irrelevant](https://ogirardot.wordpress.com/2025/02/20/object-oriented-programming-deemed-irrelevant/) _2025-02-20 · ogirardot · O. Girardot_ I ve been coding since 2006, during this time I ve seen multiple trends technologies emerge, rise and fall nowadays the elephant in the room is the bad press around OOP languages the likes of Java, C#, C++. Our profession is no stranger to this kinda of feud and debate, for example, at the start \[ \] ## [Exploratory data analysis with Scheme, Gnuplot, and Tk](https://www.travishinkelman.com/posts/eda-scheme-tk.html) _2024-12-26 · Travis Hinkelman_ In my second post on this blog , I expressed an interest in learning how to build desktop applications. I have yet to pursue that interest. Instead, I’ve primarily continued developing Shiny apps deployed on the web (but see Shiny Scorekeeper ). Recently, though, I’ve spent some time learning about the Tk GUI toolkit for developing desktop applications. In this post, I revisit an old post using… ## [Explanining Log-Sum-Exp](https://hongtaoh.com/en/2024/12/14/log-sum-exp/) _2024-12-14 · Hongtao Hao_ Motivationn & Definition In this post, we explain the concept of Log-Sum-Exp . Suppose we have three extremely small numbers: $a\_1 = e^{-200}, a\_2 = e^{-201}, a\_3 = e^{-202}$ . We are interested in the log of their sum. There are two issues to solve here: The sum will still be very small. To make the result more readible, we want to take the logarithm of it. Dealing with extremely small numbers… ## [Infinite Gaussian Mixture Model](https://hongtaoh.com/en/2024/12/12/igmm/) _2024-12-12 · Hongtao Hao_ Concept Explanation The Infinite Gaussian Mixture Model (IGMM) is an extension of the Gaussian Mixture Model . It handles situations where we don’t know how many clusters exist. There are generally two approaches to solve this problem. The first is to test different numbers of clusters and compare which result is better, using common methods like the Elbow method (using Sum of Squared… ## [Finnish Seedboxes — Up to 20Gbps (Sponsored)](https://crawlproof.com/a/HZiPaoNTLe7m) _2024-12-12 · **Sponsored**_ Private Finnish seedboxes with 1–20Gbps speeds, RAID storage and 14‑day refund ## [Reading JSON files as Scheme dataframes](https://www.travishinkelman.com/posts/json-dataframe-scheme.html) _2024-11-06 · Travis Hinkelman_ In a previous post , I wrote about reading and writing JSON files in R and Chez Scheme. After updating that post, I was curious about how much code it would take to convert a Scheme object read from a JSON file into a dataframe . It is arguably a rare circumstance that a dataframe is the best data representation for JSON. In this post, we will tackle the limited case of reading JSON files that… ## [Horse racing game win probability](https://www.travishinkelman.com/posts/horse-game.html) _2024-10-26 · Travis Hinkelman_ When gathering with family, we like to play yard, card, and board games. On a recent visit, one of the favorite games was the Across the Board Kentucky Derby Horse Racing Game . The game produced a lot of cheers and jeers and provided a fun diversion with no skill and little concentration required. After losing a little money, though, I lost interest in playing the game and decided to write some R… ## [RPI and SOS in Scheme, Python, and Elixir](https://www.travishinkelman.com/posts/rpi.html) _2024-09-02 · Travis Hinkelman_ Last spring, I played in a 3x3 basketball leage with 14 teams and only 6 regular-season games. The unbalanced schedule made me wonder if we would end up with wonky playoff seeding. I thought it would be fun to calculate the Rating Percentage Index (RPI) and Strength Of Schedule (SOS) for each team to assess discrepancies between W-L record and team rating. I was mostly following the R code for RPI…