mirror of
				https://github.com/Brandon-Rozek/website.git
				synced 2025-10-30 13:41:12 +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
		Add a link
		
	
		Reference in a new issue