What is one hot code? Why use it and when?

Click on the above “ Xiaobai studies vision ”, Optional plus " Star standard " or “ Roof placement ”

 Heavy dry goods , First time delivery 

Reading guide

When you're playing ML When modeling , You will meet this anywhere “One hot encoding” The term of .

When you're playing ML When modeling , You will meet this anywhere “One hot encoding” The term . You can see a one hot Encoder sklearn file , It says “ Use one-hot That is to say one-of- k Pattern coding classifies integer features ”. Not very clear , Right ？ Or at least not for me . Let's see one hot What is coding .

One hot The coding method is to convert the classified variables into a form , This form can be provided to ML Algorithm , In order to better predict .

Suppose the data set is as follows :

╔════════════╦═════════════════╦════════╗ 
 ║ CompanyName Categoricalvalue ║ Price  ║
 ╠════════════╬═════════════════╣════════║ 
 ║ VW         ╬      1          ║ 20000  ║
 ║ Acura      ╬      2          ║ 10011  ║
 ║ Honda      ╬      3          ║ 50000  ║
 ║ Honda      ╬      3          ║ 10000  ║
 ╚════════════╩═════════════════╩════════╝

The classification value represents the value of entries in the dataset . for example ： If there is another company in the data set , Its classification value should be 4. As the number of unique entries increases , The classification value increases accordingly .

The above table is just a representation . actually , The classification value is from 0 Start all the way to N-1 Categories .

You may already know , have access to sklearn Of LabelEncoder Complete classification value allocation .

Now let's go back to one hot code ： Let's say that we follow sklearn According to the instructions given in the document one hot code , Then do some cleaning , Finally, the following results are obtained :

╔════╦══════╦══════╦════════╦
 ║ VW ║ Acura║ Honda║ Price  ║
 ╠════╬══════╬══════╬════════╬
 ║ 1  ╬ 0    ╬ 0    ║ 20000  ║
 ║ 0  ╬ 1    ╬ 0    ║ 10011  ║
 ║ 0  ╬ 0    ╬ 1    ║ 50000  ║
 ║ 0  ╬ 0    ╬ 1    ║ 10000  ║
 ╚════╩══════╩══════╩════════╝

0 Does not exist ,1 Indicates presence .

Before we go any further , Can you think of a reason ？ Why is it not enough to use tag coding to train the model ？ Why one hot code ？

The problem with tag coding is , It assumes that the higher the category value , The better the category .“ wait , what ! ？”

Let me explain ： The premise of this form of organization is based on the value of analogy ,VW > Acura > Honda. Suppose your model calculates an average internally , So we get ,1+3 = 4/2 =2. It means ：VW and Honda The average level of is Acura. This is definitely a disaster . There will be many errors in the prediction of this model .

That's why we use one hot Encoder to execute class “ Two valued ”, And take it as a feature to train the model .

Another example ： Suppose you have a “flower” features , It's acceptable “daffodil”、“lily” and “rose” Value . One one hot Coding will “flower” The feature is transformed into three features ,“is_daffodil”、“is_lily” and “is_rose”, They are all binary .

See the picture below ：

The good news ！

Xiaobai learns visual knowledge about the planet

Open to the outside world

 download 1：OpenCV-Contrib Chinese version of extension module 

 stay 「 Xiaobai studies vision 」 Official account back office reply ： Extension module Chinese course , You can download the first copy of the whole network OpenCV Extension module tutorial Chinese version , Cover expansion module installation 、SFM Algorithm 、 Stereo vision 、 Target tracking 、 Biological vision 、 Super resolution processing and other more than 20 chapters .


 download 2：Python Visual combat project 52 speak 
 stay 「 Xiaobai studies vision 」 Official account back office reply ：Python Visual combat project , You can download, including image segmentation 、 Mask detection 、 Lane line detection 、 Vehicle count 、 Add Eyeliner 、 License plate recognition 、 Character recognition 、 Emotional tests 、 Text content extraction 、 Face recognition, etc 31 A visual combat project , Help fast school computer vision .


 download 3：OpenCV Actual project 20 speak 
 stay 「 Xiaobai studies vision 」 Official account back office reply ：OpenCV Actual project 20 speak , You can download the 20 Based on OpenCV Realization 20 A real project , Realization OpenCV Learn advanced .


 Communication group 

 Welcome to join the official account reader group to communicate with your colleagues , There are SLAM、 3 d visual 、 sensor 、 Autopilot 、 Computational photography 、 testing 、 Division 、 distinguish 、 Medical imaging 、GAN、 Wechat groups such as algorithm competition （ It will be subdivided gradually in the future ）, Please scan the following micro signal clustering , remarks ：” nickname + School / company + Research direction “, for example ：” Zhang San  +  Shanghai Jiaotong University  +  Vision SLAM“. Please note... According to the format , Otherwise, it will not pass . After successful addition, they will be invited to relevant wechat groups according to the research direction . Please do not send ads in the group , Or you'll be invited out , Thanks for your understanding ~