objet 'numpy.float64' n'est pas iterable - meanshift clustering

Question

python newbie ici. J'essaye d'exécuter ce code mais j'obtiens le message d'erreur que l'objet n'est pas itérable. J'apprécierais quelques conseils sur ce que je fais mal. Merci.

import matplotlib.pyplot as plt 
import numpy as np 
import pandas as pd 

temp = pd.read_csv("file.csv", encoding='latin-1') 
xy = temp.ix[:,2:6] 
X = xy.values 

---

X 
array([[    nan,    nan], 
    [    nan,    nan], 
    [ 3.92144000e+00,    nan], 
    [ 4.42382000e+00,    nan], 
    [ 4.18931000e+00, 5.61562775e+02], 
    [    nan,    nan], 
    [ 4.33025000e+00, 6.73123391e+02], 
    [ 6.43775000e+00,    nan], 
    [ 3.12299000e+00, 2.21886627e+03], 
    [    nan,    nan], 
    [    nan,    nan]]) 

---

from itertools import cycle 
colors = cycle('bgrcmykbgrcmykbgrcmykbgrcmyk') 

class Mean_Shift: 
    def __init__(self, radius=4): 
     self.radius = radius 

    def fit(self, data): 
     centroids = {} 

     for i in range(len(data)): 
      centroids[i] = data[i] 

     while True: 
      new_centroids = [] 
      for i in centroids: 
       in_bandwidth = [] 
       centroid = centroids[i] 
       for featureset in data: 
        if np.linalg.norm(featureset-centroid) < self.radius: 
         in_bandwidth.append(featureset) 

       new_centroid = np.average(in_bandwidth, axis=0) 
       new_centroids.append(tuple(new_centroid)) 

      uniques = sorted(list(set(new_centroids))) 

      prev_centroids = dict(centroids) 

      centroids = {} 
      for i in range(len(uniques)): 
       centroids[i] = np.array(uniques[i]) 

      optimized = True 

      for i in centroids: 
       if not np.array_equal(centroids[i], prev_centroids[i]): 
        optimized = False 

       if not optimized: 
        break 

      if optimized: 
       break 

     self.centroids = centroids 

    def predict(self,data): 
     pass 


clf = Mean_Shift() 
clf.fit(X) 

centroids = clf.centroids 

plt.scatter(X[:,0],X[:,1],s=50) 

for c in centroids: 
plt.scatter(centroids[c][0], centroids[c][1], color = 'k', marker='*', s=150) 

plt.show() 

Voici le code d'erreur que je reçois:

/Users/carla/anaconda/lib/python3.5/site-packages/numpy/core/_methods.py:59: RuntimeWarning: Mean of empty slice. 
    warnings.warn("Mean of empty slice.", RuntimeWarning) 
--------------------------------------------------------------------------- 
TypeError         Traceback (most recent call last) 
<ipython-input-11-e13932b6e72d> in <module>() 
50 
51 clf = Mean_Shift() 
---> 52 clf.fit(X) 
53 
54 centroids = clf.centroids 

<ipython-input-11-e13932b6e72d> in fit(self, data) 
22 
23     new_centroid = np.average(in_bandwidth, axis=0) 
---> 24     new_centroids.append(tuple(new_centroid)) 
25 
26    uniques = sorted(list(set(new_centroids))) 

TypeError: 'numpy.float64' object is not iterable 

Answer 1

new_centroid = np.average(in_bandwidth, axis=0) 

est d'attribuer un scalaire à new_centroid alors vous êtes TRYI ng à tuple(scalar) qui lance l'erreur.

tuple(2.) 

--------------------------------------------------------------------------- 
TypeError         Traceback (most recent call last) 
<ipython-input-51-4406f9e676cf> in <module>() 
----> 1 tuple(2.) 

TypeError: 'float' object is not iterable