mlpack_prepro..._hot_encoding(1)

✨ NAME ✨

mlpack_preprocess_one_hot_encoding — one hot encoding, bestie 💅

So basically this whole manual page lives under the name mlpack_preprocess_one_hot_encoding in the mlpack_prepro..._hot_encoding(1) fam, and it's here to make your categorical data go from "meh" to main character. It does one hot encoding. That's the whole vibe. 🎀

---

🦋 SYNOPSIS 🦋

mlpack_preprocess_one_hot_encoding -i string [-d vector] [-V bool] [-o unknown] [-h -v]

That's the shape of it. Every bracket-y bit is optional, everything else is not. Slay.

---

💖 DESCRIPTION 💖

okay so picture this: you hand this utility a dataset, plus a vector of indices, and it goes ahead and does one-hot encoding on the features sitting at exactly those indices. Indices = the IDs of the dimensions you want one-hot encoded, periodt.

But here's the twist 👀 — if you don't specify any dimensions with --dimensions (-d), then all the categorical-type dimensions get one-hot encoded, no questions asked. If you do give it some dimensions with --dimensions (-d), then only those ones get the glow-up. Not a single extra. 🧚

Your freshly-encoded matrix can be saved off with the --output_file (-o) parameters. Because what's the point of a transformation you can't keep, right?

So, a simple example where we want to encode 1st and 3rd feature from dataset 'X.csv' into 'X_output.csv' would be

$ mlpack_preprocess_one_hot_encoding --input_file X.arff --output_file X_output.csv --dimensions 1 --dimensions 3

(Yes bestie, the prose says X.csv and the command says X.arff. That's just how the original page wrote it. Reproducing it faithfully, no cap. 🙈)

---

🔥 REQUIRED INPUT OPTIONS 🔥

Like, you genuinely cannot skip this one. It's giving non-negotiable.

• --input_file (-i) [string] — Matrix containing data. That's your raw material, your canvas, your whole starting point. 💜

---

🌸 OPTIONAL INPUT OPTIONS 🌸

The little extras. The accessories to the outfit. 👑

• --dimensions (-d) [vector] — Index of dimensions that need to be one-hot encoded (if unspecified, all categorical dimensions are one-hot encoded). Default value [].
• --help (-h) [bool] — Default help info. When you're lost and just need someone to hold your hand. 🫶
• --info [string] — Print help on a specific option. Default value ''.
• --verbose (-v) [bool] — Display informational messages and the full list of parameters and timers at the end of execution. It's the "tell me everything" flag. ⭐
• --version (-V) [bool] — Display the version of mlpack. Which mlpack are you even running? Find out. ✨

---

🎀 OPTIONAL OUTPUT OPTIONS 🎀

Where the results live once the magic has happened.

• --output_file (-o) [unknown] — Matrix to save one-hot encoded features data to. Your encoded data, saved, permanent, iconic. 💅

---

🌈 ADDITIONAL INFORMATION 🌈

For further information, including relevant papers, citations, and theory, consult the documentation found at http://www.mlpack.org or included with your distribution of mlpack. iykyk — if you want the deep lore, the theory, the citations, that's where you go. 📚

---

💀 Quick reminder bestie: if you skip --dimensions (-d), it's every categorical dimension, not just some. Be sure that's what you want before you run it!

---

that's the whole page, and honestly? we ate. This gem is shipped as mlpack-4.8.0, dated 22 June 2026, under the manual-page name mlpack_prepro..._hot_encoding(1). byeee 🦋✨