CODEWITHSUNDEEP
pythonscikit-learn
Sumanth
Sumanth

Reputation: 383

Facing Issue with the decision tree classifier implementation in scikit learn

Trying to generate a decision tree in sci-kit learn. I have a CSV file, providing as input to my sci-kit program. When I print the dataset length it is 502, the data set shape is (502, 1).There is only one array.

How do I fit into the decision tree and get a result, not sure if I am doing it correctly, below is my code. 

    import numpy as np
    import pandas as pd
    from sklearn import tree
    from sklearn.cross_validation import train_test_split
    from sklearn.tree import DecisionTreeClassifier
    from sklearn.metrics import accuracy_score

    input_file = "output.csv"

    # for tab delimited use:

     df = pd.read_csv(input_file, header = 0, delimiter = "\t")

   # printing the original column values in a python list

     print(df.values)

     print("DataSet Length :",len(df))

     print("DataSet Shape :",df.shape)

    # Assigning values to an array  
     X=df.values[:,0]

   # test train the the data
     X_train,X_test=train_test_split(X,test_size=0.3,random_state=100)

   # Passing to the Decision Tree Classifier, with entropy criterion

    clf_entropy = DecisionTreeClassifier(criterion = "entropy", rando  
    m_state = 100,max_depth=3, min_samples_leaf=5)

    # Fitting the data  to the classifier
    clf_entropy.fit(X_train)

CSV file is on the below link

https://drive.google.com/file/d/0B3XlF206d5UrVnh6QS1LRW0xT0U/view?usp=sharing

Download and open using excel. Referring to the following sci-kit documentation for reference.

http://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier

Upvotes: 0

Views: 2538

Answers (1)

Miriam Farber
Miriam Farber

Reputation: 19634

In order to fit a decision tree classifier, your training and testing data needs to have labels. Using these labels, you can fit the tree. Here is an example from sklearn website:

from sklearn import tree
X = [[0, 0], [1, 1]]
Y = [0, 1]
clf = tree.DecisionTreeClassifier()
clf = clf.fit(X, Y)

The problem is that in your code, you have only X values, without labels (Y values). So you cannot fit the tree.

Upvotes: 2

Related Questions