"ValueError: setting an array element with a sequence" during scikit-learn classification

Using the tutorial on multiclass adaboost, I'm trying to classify some images that have two classes (but I don't suppose the algorithm shouldn't work if the problem is binary). Then I'm going to extend my samples to include other classes.

My current test is quite small, only 17 images in all, 10 for training, 7 for testing. (I'll add more samples later, right now I just want to get the code up and running).

For now I have two classes: 0: no vehicle, 1: vehicle present I used integer labels because according to the example in the link above, the training data consists of integer-based labels.

I've edited the provided example only a bit, to include my own image files, but I'm getting the following error:

Traceback (most recent call last):
  File "C:\Users\app\Documents\Python Scripts\carclassify.py", line 66, in <module>
    bdt_discrete.fit(X_train, y_train)
  File "C:\Users\app\Anaconda\lib\site-packages\sklearn\ensemble\weight_boosting.py", line 389, in fit
    return super(AdaBoostClassifier, self).fit(X, y, sample_weight)
  File "C:\Users\app\Anaconda\lib\site-packages\sklearn\ensemble\weight_boosting.py", line 99, in fit
    X = np.ascontiguousarray(array2d(X), dtype=DTYPE)
  File "C:\Users\app\Anaconda\lib\site-packages\numpy\core\numeric.py", line 408, in ascontiguousarray
    return array(a, dtype, copy=False, order='C', ndmin=1)
ValueError: setting an array element with a sequence.

The relevant part of my code is below:

f = open("PATH_TO_SAMPLES\\samples.txt",'r')
out = f.read().splitlines()
import numpy as np

imgs = []
tmp_hogs = []
# 13 of the images are with vehicles, 4 are without
labels = [1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0]

for file in out:
        filepath = "C:\PATH_TO_SAMPLE_IMAGES\\" + file
        curr_img = color.rgb2gray(io.imread(filepath))
        imgs.append(resize(curr_img,(60,40)))
        fd, hog_image = hog(curr_img, orientations=8, pixels_per_cell=(16, 16),
                 cells_per_block=(1, 1), visualise=True)
        tmp_hogs.append(fd)

img_hogs = np.array(tmp_hogs)
n_split = 10
X_train, X_test = img_hogs[:n_split], X[n_split:] # all first ten images with vehicles
y_train, y_test = labels[:n_split], labels[n_split:] # 3 images with vehicles, 4 without

#now all the code below is straight off the example on scikit-learn's website

bdt_real = AdaBoostClassifier(
    DecisionTreeClassifier(max_depth=2),
    n_estimators=600,
    learning_rate=1)

bdt_discrete = AdaBoostClassifier(
    DecisionTreeClassifier(max_depth=2),
    n_estimators=600,
    learning_rate=1.5,
    algorithm="SAMME")

bdt_real.fit(X_train, y_train)
bdt_discrete.fit(X_train, y_train)

real_test_errors = []
discrete_test_errors = []

for real_test_predict, discrete_train_predict in zip(
        bdt_real.staged_predict(X_test), bdt_discrete.staged_predict(X_test)):
    real_test_errors.append(
        1. - accuracy_score(real_test_predict, y_test))
    discrete_test_errors.append(
        1. - accuracy_score(discrete_train_predict, y_test))

n_trees = xrange(1, len(bdt_discrete) + 1)

What am I doing wrong, and what needs to be fixed?

Edit

Here are the samples, in order to replicate the error:

1. positive sample 1

2. positive sample 2

3. positive sample 3

4. positive sample 4

5. positive sample 5

6. positive sample 6

7. positive sample 7

8. positive sample 8

9. positive sample 9

10. positive sample 10

11. positive sample 11

12. positive sample 12

13. positive sample 13

14. negative sample 1

15. negative sample 2

16. negative sample 3

17. negative sample 4