mirror of
https://github.com/Brandon-Rozek/website.git
synced 2025-05-31 23:04:44 +00:00
New Post
This commit is contained in:
parent
f3d1c25d3b
commit
e63bbffaac
1 changed files with 22 additions and 0 deletions
22
content/blog/reverseonehotencode.md
Normal file
22
content/blog/reverseonehotencode.md
Normal file
|
@ -0,0 +1,22 @@
|
||||||
|
---
|
||||||
|
title: "Reverse One-Hot Encode"
|
||||||
|
date: 2020-10-11T21:58:47-04:00
|
||||||
|
draft: false
|
||||||
|
tags: []
|
||||||
|
---
|
||||||
|
|
||||||
|
Let's say that you have a dataset that is one hot encoded like the following observation:
|
||||||
|
|
||||||
|
```python
|
||||||
|
import numpy as np
|
||||||
|
obs = np.array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0,
|
||||||
|
0, 0, 0, 0])
|
||||||
|
```
|
||||||
|
|
||||||
|
The easiest way to reverse one-hot encode the structure, is to take the `argmax` of the observation.
|
||||||
|
|
||||||
|
```python
|
||||||
|
reverse_encoding = np.argmax(obs)
|
||||||
|
# 13
|
||||||
|
```
|
||||||
|
|
Loading…
Add table
Reference in a new issue