ValueError：无法将字符串转换为浮点数：'f'

Question

Want to use Decision Tree algorithm. But getting some error as ValueError: could not convert string to float: 'f'. Kindly help me where am I missing.

import numpy as np
import pandas as pd

#Plotting
import matplotlib.pyplot as plt

#Machine Learning Libraries

from sklearn.neighbors import KNeighborsClassifier
from sklearn import tree   
from sklearn.model_selection import train_test_split

from sklearn import metrics
from sklearn.metrics import accuracy_score
from sklearn.metrics import confusion_matrix



#Loading Dataset
My_dataset = pd.read_csv('mushrooms.csv');
print (My_dataset.head())
print (My_dataset.shape)
#Dividing the datasets into Indicator and Predictor Variables
My_data = My_dataset.iloc[:,1:23].values
My_target = My_dataset.iloc[:,0].values
print()
print(My_data)
print()
print(My_target)
mushroom_train,mushroom_test,mushroomtarget_train,mushroomtarget_test = \
train_test_split(My_data,My_target, test_size = 0.3)

DT_Model_Mushroom = tree.DecisionTreeClassifier() 
DT_Model_Mushroom_Fitted = DT_Model_Mushroom.fit(mushroom_train, mushroomtarget_train)

Error:

Error
return array(a, dtype, copy=False, order=order)

ValueError: could not convert string to float: 'f'

Answer 1

The Decision Tree Classifier in scikit learn does not take Strings as input.

If you have categorical variables in your data, you should encode them before (with one of the sklearn encoder for example : One hot encoder , Ordinal Encoder , ... )

If you don't have categorical variables in your data, pandas may not be able to correctly attribute types to your columns. If this situation you should use the "dtype" argument of the read_csv function.