1

This is the lines:

from sklearn import tree


X = [[181,80,44], [177, 70, 43], [160, 60, 38], 154, 54, 37],
[166,64,40], [190,90,47], [175,64,39],[177,70,40],[159,55,37],
[171,75,42],[181,85,43]

Y = ['male', 'female', 'female', 'female', 'male', 'male', 'male',
'female', 'male', 'female', 'male']

clf = tree.DecisionTreeClassifier()

clf = clf.fit(X,Y)

prediction = clf.predict([[182,78,43]])

print (prediction)

Result:

Traceback (most recent call last):


File "C:\Python\code\test.py", line 14, in <module>
clf = clf.fit(X,Y)
File "C:\Python\lib\site-packages\sklearn\tree\tree.py", line 790, in fit
X_idx_sorted=X_idx_sorted)
File "C:\Python\lib\site-packages\sklearn\tree\tree.py", line 116, in fit
X = check_array(X, dtype=DTYPE, accept_sparse="csc")
File "C:\Python\lib\site-packages\sklearn\utils\validation.py", line 402, 

in check_array
array = np.array(array, dtype=dtype, order=order, copy=copy)
in check_array
array = np.array(array, dtype=dtype, order=order, copy=copy)
ValueError: setting an array element with a sequence.
[Finished in 0.5s]

Expected result:

Should display gender predicted from body measurements: "182,78,43"

Example: male or female

Run with Python 3.6 using sklearn, numpy+mkl and scipy on Sublime.

The code is originally from: https://www.youtube.com/watch?v=T5pRlIbr6gg. No answer in the whole youtube comment section to this. Appreciate if answer can be found here, could not find any answer.

Improve this question
0

1 Answer 1

Reset to default
1

If that's really your code, the problem starts right at the beginning:

X = [[181,80,44], [177, 70, 43], [160, 60, 38], 154, 54, 37], 
[166,64,40], [190,90,47], [175,64,39],[177,70,40],[159,55,37],
[171,75,42],[181,85,43]

is creating something far far away from being usable:

print(X)
# ([[181, 80, 44], [177, 70, 43], [160, 60, 38], 154, 54, 37],)

So there is one bracket of each orientation to add (if that's not clear to you: read sklearn's docs on the data-format: 2d-array of shape (n_samples, n_features); consider also reading some introduction to numpy where the word shape comes from -> internally everything is numpy-based):

X = [[181,80,44], [177, 70, 43], [160, 60, 38], [154, 54, 37],  # before 154
[166,64,40], [190,90,47], [175,64,39],[177,70,40],[159,55,37],
[171,75,42],[181,85,43]]                                        # at end

I have to admit: that's something which should be found immediately and i can't understand why someone would invest time to create a post for SO, but not invest time to check the syntax of a simple array-creation.

To be fair: i thought it would not be going through the syntax-check at first (it's really a strange construction as stated).

Edit: To be fair #2: it's really that bad in that linked video too... I'm not sure what to think of that (well using a DTree for this task is probably already nuts, even LinearRegression seems more viable)!

And yes, the code predicts male after the correction has been done as above!

Improve this answer
Sign up to request clarification or add additional context in comments.

4 Comments

The initial approach to begin with was as I could've imagined, like the dependancies (sklearn) as explained in the video required to have a firm ground to work upon and something to retrieve from is easily related to this phenomenon as well. Thank you for being able to be concise and detailed in your statement. I've learned from this!
Make sure to grab the basics: numpy, sklearn. It's important. I don't consider this video-channel high-quality at first sight (broken code, interesting algo-decision for this prob, presentation and especially the about-page)! I have to admit: without checking out python's docs, i'm not really understanding what exactly your X is (outer structure is a tuple), but it does not matter as it's not following the assumptions needed.
Out of the vast tutorials, documents etc available to learn the basics, any recommendations to begin with for an autodidact? As for the second expression, learning backwards is the answer to that.
I don't know what you want to learn. But understanding the theory and then understanding the technical-stuff will suffice. That means: look for open university ML-courses, and then follow the official numpy/scipy/sklearn docs.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.