Left: 32-bit Excel converting text to numbers; Right: xgboost running in R

CPU Usage Comparison

Converting text to numbers in Excel versus XGBoost in R

Once in a while, Excel will take a ridiculous amount of time trying to convert numbers stored as text to actual numbers. Why this happens, I have no idea. My best guess is that when those ‘numbers stored as text’ were pasted into an Excel worksheet, some other info got pasted in too.

When this happens, my computer freezes up and I have to wait up to 30 minutes for Excel to calm down. In the situations where I decide it’s more worthwhile to force close Excel and lose any unsaved work, I can usually get around the freeze by multiplying the ‘numbers stored as text’ by 1 or by using the text-to-columns function and getting rid of all delimiters.

To my simple humanoid mind, converting text to numbers (concept change) seems way easier than running through the xgboost algorithm (optimizing formidable math equations). The computer processing power required for these two processes? I have no idea.

Anyways, I found this to be pretty nifty.

Like what you read? Give JJ a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.