Skip to content

Instantly share code, notes, and snippets.

View schnee's full-sized avatar

Brent Schneeman schnee

View GitHub Profile
@schnee
schnee / gist:30e4e9f06f4adbe8bc2e937a419864db
Last active August 13, 2023 17:45
Astro processing notes
Seems to work with Milky Way images. But I'm not sure what to do with it
define sequence
convert sequence to FITS
stack
remove background
plate solve withe the belot
solve-field --scale-low 45 --scale-units degwidth --crpix-center --overwrite --downsample 2 r_saggita_stacked_crop.fit
@schnee
schnee / purrr_happy.R
Created May 5, 2019 18:08
What considerations are important in choosing between the two styles below? Interpretability, maintainability, speed, other?
library(purrr)
y_test <- sample(0:9, 2000, replace = TRUE)
pred1 <- sample(0:9, 2000, replace= TRUE)
pred2 <- sample(0:9, 2000, replace= TRUE)
preds <- list(pred1, pred2)
# want to apply the following to preds
# sum(diag(table(y_test, pred)))/length(y_test)
@schnee
schnee / gist:d9333f4a6e2d8215064520ea820fded4
Created December 30, 2017 15:54
history for hugin setup
ls
pwd
hi
man hi
man cats
yes "we have no bananas"
yes
man yes
exit
git
// Databricks notebook source exported at Sat, 6 Aug 2016 14:28:53 UTC
// MAGIC %md
// MAGIC # Breckenridge Property Description Topic Modeling
// MAGIC This notebook turns the text contained in property descriptions in the Breckenridge CO US destination into topic probability distributions for subsequent math. The chief output is the LDA-determined topic distributions. These distributions are analyzed for similarity scores in an R document elsewhere.
// MAGIC
// MAGIC For getting all the topic distributions, you'll want the 'clusteredDF' object
// COMMAND ----------
import org.apache.spark.sql.SaveMode;