如何在 Python 中进行一次热编码并以编程方式获取类数？ 对于人工神经网络

Question

I have a list called y_train that looks like this

[   307    307    307 ... 257947 257947 257947]

The 307, 257947 are all distinct IDs that I want to do one hot encoding on. It has a total of 480 classes. The list y_train has a length of 10799 also as the number of rows. How can I do one hot encoding on this so that it shows that it has 480 class and 10799 rows? I'm trying to fit this in a tensorflow model.

Answer 1

There are multiple ways to do this. One of the methods is to use pandas.get_dummies to convert categorical variables into dummy/indicator variables.

Example:

df = pd.DataFrame({'A': ['a', 'b', 'a'], 'B': ['b', 'a', 'c'],
                   'C': [1, 2, 3]})

pd.get_dummies(df, prefix=['col1', 'col2'])
   C  col1_a  col1_b  col2_a  col2_b  col2_c
0  1       1       0       0       1       0
1  2       0       1       1       0       0
2  3       1       0       0       0       1

Another method is to use Scikit-learn's OneHotEncoder