Yorgos
LightNews LightNews
banner
gd047.bsky.social
Yorgos
@gd047.bsky.social
45 followers 290 following 92 posts
Am Heimcomputer sitze ich hier und programmieR die Zukunft mir.
Posts Replies Media Videos
gd047.bsky.social
The Plotly version of the graph.
July 28, 2025 at 1:07 PM
gd047.bsky.social
Patent Alert: The Sperm Plot™ 😂 🧬📊
Novel visualization where project timelines swim upstream toward completion. Each 'swimmer' tracks % progress vs % time elapsed.

#DataViz #Analytics #PatentPending #RStats
July 26, 2025 at 8:50 AM
gd047.bsky.social
April 11, 2025 at 5:19 PM
gd047.bsky.social
"Python can do everything R can do!*

*Terms and conditions apply: may actually be running R code behind the scenes 😂
#Rstats #PythonVsR #Statistics
March 5, 2025 at 7:11 AM
gd047.bsky.social
I really liked the graph in Figure 5 here:
`Low-density lipoprotein and subclinical atherosclerosis`
www.jacc.org/doi/10.1016/...
So I tried to simulate it using ggplot. #rstats
gist.github.com/gd047/73e23a...
January 24, 2025 at 12:39 PM
gd047.bsky.social
Buildings account for nearly 40% of primary energy consumption. Advanced #MachineLearning and statistical methods are transforming how we model, predict, and optimize heating and cooling energy use. 🔽

#BuildingEnergy #EnergyEfficiency #BuildingSimulation #Sustainability
January 19, 2025 at 4:31 PM
gd047.bsky.social
Feeling very satisfied to have completed my {targets}-powered R pipeline for efficient report generation (with a little help from A.I. 😜)!

Effortless automation, faster results, and peace of mind! 🎉
#rstats #automation #reportgeneration

Here's how it works:
January 19, 2025 at 7:32 AM
gd047.bsky.social
🤔 Spent hours figuring out why my dfs were losing their col types when passed as params to .qmd. The issue: Quarto's serialization converts everything to chr columns! 😤 The solution? pass the file paths instead and let Quarto load the data itself 🎯
Is this documented anywhere? #rstats #quarto
January 14, 2025 at 9:53 AM
gd047.bsky.social
I have no experience. I've worked a bit with {targets} and I like it. I just read that it's a simpler version of it and I was wondering if I could use it as a more appropriate solution -> "...used when up-to-dateness of the source data is unknown (e.g., coming from an API or database)"
January 3, 2025 at 6:30 PM
gd047.bsky.social
Dieses Jahr war es für meinen Sohn einfach, ein Geschenk für mich auszusuchen. #kraftwerk #musiksky #nowspinning
January 1, 2025 at 3:30 PM
gd047.bsky.social
December 21, 2024 at 7:03 PM
gd047.bsky.social
Javascript for #rstats Shiny cheatsheet.
December 14, 2024 at 3:41 PM
gd047.bsky.social
I just completed "Bridge Repair" - Day 7 - Advent of Code 2024 #AdventOfCode adventofcode.com/2024/day/7
library(stringr) # Recursive evaluator for a given target and numbers evaluate_recursively <- function(test_value, numbers, index = 2, current_value = numbers[1]) { # Base case: If we've processed all numbers, check if the result matches the target if (index > length(numbers)) return(!is.na(current_value) && current_value == test_value) # Explore "+" operation if (evaluate_recursively(test_value, numbers, index + 1, current_value + numbers[index])) return(TRUE) # Explore "*" operation if (evaluate_recursively(test_value, numbers, index + 1, current_value * numbers[index])) return(TRUE) # Explore "||" (concatenation) operation new_value <- as.numeric(paste0(current_value, numbers[index])) if (!is.na(new_value) && evaluate_recursively(test_value, numbers, index + 1, new_value)) return(TRUE) # If none of the operations lead to the target, return FALSE FALSE } # Process a single line process_line <- function(line) { # Split the line into target and numbers parts <- strsplit(line, ": ")[[1]] if (length(parts) != 2) return(0) # Skip malformed lines target <- as.numeric(parts[1]) numbers <- as.numeric(strsplit(parts[2], "\\s+")[[1]]) # Evaluate the line using recursion if (evaluate_recursively(target, numbers)) return(target) 0 } # Main function to process all lines in the file process_file_recursive <- function(file_path) { lines <- readLines(file_path) results <- sapply(lines, process_line) sum(results) } input_file <- "input07.txt" process_file_recursive(input_file)
December 7, 2024 at 7:10 PM
gd047.bsky.social
Day 7 of #adventofcode in #rstats (Part 1)
library(dplyr) library(stringr) evaluate_left_to_right <- function(numbers, symbols) { result <- numbers[1] for (i in seq_along(symbols)) { if (symbols[i] == "+") { result <- result + numbers[i + 1] } else if (symbols[i] == "*") { result <- result * numbers[i + 1] } } result } process_file <- function(file_path) { lines <- readLines(file_path) # Pre-generate symbol combinations (assuming max ~10 numbers for practicality) max_symbols <- max(str_count(lines, "\\s+") + 1) - 1 all_symbols <- expand.grid(rep(list(c("+", "*")), max_symbols)) all_symbols <- as.matrix(all_symbols) results <- vector("numeric", length(lines)) for (i in seq_along(lines)) { line <- lines[i] parts <- strsplit(line, ": ")[[1]] if (length(parts) != 2) { results[i] <- 0 next } target <- as.numeric(parts[1]) numbers <- as.numeric(strsplit(parts[2], "\\s+")[[1]]) n <- length(numbers) # Filter valid symbol combinations valid_symbols <- all_symbols[1:(2^(n - 1)), 1:(n - 1)] # Compute results for all expressions expr_results <- apply(valid_symbols, 1, evaluate_left_to_right, numbers = numbers) results[i] <- if (target %in% expr_results) target else 0 } sum(results) } # Apply function to the input file input_file <- "input07.txt" process_file(input_file)
December 7, 2024 at 1:00 PM
gd047.bsky.social
A page from Srinivasha Ramanujan's 1913 letter to mathematician G. H. Hardy, presenting mathematical relationships and formulas he discovered.
November 22, 2024 at 7:30 AM