Deep Learning

Assignment 1

The objective of this assignment is to learn about simple data curation practices, and familiarize you with some of the data we'll be reusing later.

This notebook uses the notMNIST dataset to be used with python experiments. This dataset is designed to look like the classic MNIST dataset, while looking a little more like real data: it's a harder task, and the data is a lot less 'clean' than MNIST.


In [2]:
# These are all the modules we'll be using later. Make sure you can import them
# before proceeding further.
from __future__ import print_function
import matplotlib.pyplot as plt
import numpy as np
import os
import sys
import tarfile
from IPython.display import display, Image
from scipy import ndimage
from sklearn.linear_model import LogisticRegression
from six.moves.urllib.request import urlretrieve
from six.moves import cPickle as pickle

# Config the matlotlib backend as plotting inline in IPython
%matplotlib inline


/usr/local/lib/python3.4/dist-packages/matplotlib/backends/backend_gtk3agg.py:18: UserWarning: The Gtk3Agg backend is known to not work on Python 3.x with pycairo. Try installing cairocffi.
  "The Gtk3Agg backend is known to not work on Python 3.x with pycairo. "

First, we'll download the dataset to our local machine. The data consists of characters rendered in a variety of fonts on a 28x28 image. The labels are limited to 'A' through 'J' (10 classes). The training set has about 500k and the testset 19000 labelled examples. Given these sizes, it should be possible to train models quickly on any machine.


In [3]:
url = 'http://commondatastorage.googleapis.com/books1000/'
last_percent_reported = None

def download_progress_hook(count, blockSize, totalSize):
  """A hook to report the progress of a download. This is mostly intended for users with
  slow internet connections. Reports every 1% change in download progress.
  """
  global last_percent_reported
  percent = int(count * blockSize * 100 / totalSize)

  if last_percent_reported != percent:
    if percent % 5 == 0:
      sys.stdout.write("%s%%" % percent)
      sys.stdout.flush()
    else:
      sys.stdout.write(".")
      sys.stdout.flush()
      
    last_percent_reported = percent
        
def maybe_download(filename, expected_bytes, force=False):
  """Download a file if not present, and make sure it's the right size."""
  if force or not os.path.exists(filename):
    print('Attempting to download:', filename) 
    filename, _ = urlretrieve(url + filename, filename, reporthook=download_progress_hook)
    print('\nDownload Complete!')
  statinfo = os.stat(filename)
  if statinfo.st_size == expected_bytes:
    print('Found and verified', filename)
  else:
    raise Exception(
      'Failed to verify ' + filename + '. Can you get to it with a browser?')
  return filename

train_filename = maybe_download('notMNIST_large.tar.gz', 247336696)
test_filename = maybe_download('notMNIST_small.tar.gz', 8458043)


Found and verified notMNIST_large.tar.gz
Found and verified notMNIST_small.tar.gz

Extract the dataset from the compressed .tar.gz file. This should give you a set of directories, labelled A through J.


In [4]:
num_classes = 10
np.random.seed(133)

def maybe_extract(filename, force=False):
  root = os.path.splitext(os.path.splitext(filename)[0])[0]  # remove .tar.gz
  if os.path.isdir(root) and not force:
    # You may override by setting force=True.
    print('%s already present - Skipping extraction of %s.' % (root, filename))
  else:
    print('Extracting data for %s. This may take a while. Please wait.' % root)
    tar = tarfile.open(filename)
    sys.stdout.flush()
    tar.extractall()
    tar.close()
  data_folders = [
    os.path.join(root, d) for d in sorted(os.listdir(root))
    if os.path.isdir(os.path.join(root, d))]
  if len(data_folders) != num_classes:
    raise Exception(
      'Expected %d folders, one per class. Found %d instead.' % (
        num_classes, len(data_folders)))
  print(data_folders)
  return data_folders
  
train_folders = maybe_extract(train_filename)
test_folders = maybe_extract(test_filename)


notMNIST_large already present - Skipping extraction of notMNIST_large.tar.gz.
['notMNIST_large/A', 'notMNIST_large/B', 'notMNIST_large/C', 'notMNIST_large/D', 'notMNIST_large/E', 'notMNIST_large/F', 'notMNIST_large/G', 'notMNIST_large/H', 'notMNIST_large/I', 'notMNIST_large/J']
notMNIST_small already present - Skipping extraction of notMNIST_small.tar.gz.
['notMNIST_small/A', 'notMNIST_small/B', 'notMNIST_small/C', 'notMNIST_small/D', 'notMNIST_small/E', 'notMNIST_small/F', 'notMNIST_small/G', 'notMNIST_small/H', 'notMNIST_small/I', 'notMNIST_small/J']

Problem 1

Let's take a peek at some of the data to make sure it looks sensible. Each exemplar should be an image of a character A through J rendered in a different font. Display a sample of the images that we just downloaded. Hint: you can use the package IPython.display.


eog's good enough for me.

Now let's load the data in a more manageable format. Since, depending on your computer setup you might not be able to fit it all in memory, we'll load each class into a separate dataset, store them on disk and curate them independently. Later we'll merge them into a single dataset of manageable size.

We'll convert the entire dataset into a 3D array (image index, x, y) of floating point values, normalized to have approximately zero mean and standard deviation ~0.5 to make training easier down the road.

A few images might not be readable, we'll just skip them.


In [5]:
image_size = 28  # Pixel width and height.
pixel_depth = 255.0  # Number of levels per pixel.

def load_letter(folder, min_num_images):
  """Load the data for a single letter label."""
  image_files = os.listdir(folder)
  dataset = np.ndarray(shape=(len(image_files), image_size, image_size),
                         dtype=np.float32)
  print(folder)
  num_images = 0
  for image in image_files:
    image_file = os.path.join(folder, image)
    try:
      # ndimage is from scipy
      # the data is also normalized here
      image_data = (ndimage.imread(image_file).astype(float) - 
                    pixel_depth / 2) / pixel_depth
      if image_data.shape != (image_size, image_size):
        raise Exception('Unexpected image shape: %s' % str(image_data.shape))
      dataset[num_images, :, :] = image_data
      num_images = num_images + 1
    except IOError as e:
      print('Could not read:', image_file, ':', e, '- it\'s ok, skipping.')
    
  dataset = dataset[0:num_images, :, :]
  if num_images < min_num_images:
    raise Exception('Many fewer images than expected: %d < %d' %
                    (num_images, min_num_images))
    
  print('Full dataset tensor:', dataset.shape)
  print('Mean:', np.mean(dataset))
  print('Standard deviation:', np.std(dataset))
  return dataset
        
def maybe_pickle(data_folders, min_num_images_per_class, force=False):
  dataset_names = []
  for folder in data_folders:
    set_filename = folder + '.pickle'
    dataset_names.append(set_filename)
    if os.path.exists(set_filename) and not force:
      # You may override by setting force=True.
      print('%s already present - Skipping pickling.' % set_filename)
    else:
      print('Pickling %s.' % set_filename)
      dataset = load_letter(folder, min_num_images_per_class)
      try:
        with open(set_filename, 'wb') as f:
          pickle.dump(dataset, f, pickle.HIGHEST_PROTOCOL)
      except Exception as e:
        print('Unable to save data to', set_filename, ':', e)
  
  return dataset_names

train_datasets = maybe_pickle(train_folders, 45000)
test_datasets = maybe_pickle(test_folders, 1800)


Pickling notMNIST_large/A.pickle.
notMNIST_large/A
Could not read: notMNIST_large/A/Um9tYW5hIEJvbGQucGZi.png : cannot identify image file - it's ok, skipping.
Could not read: notMNIST_large/A/SG90IE11c3RhcmQgQlROIFBvc3Rlci50dGY=.png : cannot identify image file - it's ok, skipping.
Could not read: notMNIST_large/A/RnJlaWdodERpc3BCb29rSXRhbGljLnR0Zg==.png : cannot identify image file - it's ok, skipping.
Full dataset tensor: (52909, 28, 28)
Mean: -0.12825
Standard deviation: 0.443121
Pickling notMNIST_large/B.pickle.
notMNIST_large/B
Could not read: notMNIST_large/B/TmlraXNFRi1TZW1pQm9sZEl0YWxpYy5vdGY=.png : cannot identify image file - it's ok, skipping.
Full dataset tensor: (52911, 28, 28)
Mean: -0.00756305
Standard deviation: 0.454492
Pickling notMNIST_large/C.pickle.
notMNIST_large/C
Full dataset tensor: (52912, 28, 28)
Mean: -0.142258
Standard deviation: 0.439806
Pickling notMNIST_large/D.pickle.
notMNIST_large/D
Could not read: notMNIST_large/D/VHJhbnNpdCBCb2xkLnR0Zg==.png : cannot identify image file - it's ok, skipping.
Full dataset tensor: (52911, 28, 28)
Mean: -0.0573678
Standard deviation: 0.455648
Pickling notMNIST_large/E.pickle.
notMNIST_large/E
Full dataset tensor: (52912, 28, 28)
Mean: -0.069899
Standard deviation: 0.452942
Pickling notMNIST_large/F.pickle.
notMNIST_large/F
Full dataset tensor: (52912, 28, 28)
Mean: -0.125584
Standard deviation: 0.447089
Pickling notMNIST_large/G.pickle.
notMNIST_large/G
Full dataset tensor: (52912, 28, 28)
Mean: -0.0945816
Standard deviation: 0.44624
Pickling notMNIST_large/H.pickle.
notMNIST_large/H
Full dataset tensor: (52912, 28, 28)
Mean: -0.0685222
Standard deviation: 0.454232
Pickling notMNIST_large/I.pickle.
notMNIST_large/I
Full dataset tensor: (52912, 28, 28)
Mean: 0.0307862
Standard deviation: 0.468898
Pickling notMNIST_large/J.pickle.
notMNIST_large/J
Full dataset tensor: (52911, 28, 28)
Mean: -0.153358
Standard deviation: 0.443657
Pickling notMNIST_small/A.pickle.
notMNIST_small/A
Could not read: notMNIST_small/A/RGVtb2NyYXRpY2FCb2xkT2xkc3R5bGUgQm9sZC50dGY=.png : cannot identify image file - it's ok, skipping.
Full dataset tensor: (1872, 28, 28)
Mean: -0.132626
Standard deviation: 0.445128
Pickling notMNIST_small/B.pickle.
notMNIST_small/B
Full dataset tensor: (1873, 28, 28)
Mean: 0.00535608
Standard deviation: 0.457115
Pickling notMNIST_small/C.pickle.
notMNIST_small/C
Full dataset tensor: (1873, 28, 28)
Mean: -0.141521
Standard deviation: 0.44269
Pickling notMNIST_small/D.pickle.
notMNIST_small/D
Full dataset tensor: (1873, 28, 28)
Mean: -0.0492167
Standard deviation: 0.459759
Pickling notMNIST_small/E.pickle.
notMNIST_small/E
Full dataset tensor: (1873, 28, 28)
Mean: -0.0599148
Standard deviation: 0.45735
Pickling notMNIST_small/F.pickle.
notMNIST_small/F
Could not read: notMNIST_small/F/Q3Jvc3NvdmVyIEJvbGRPYmxpcXVlLnR0Zg==.png : cannot identify image file - it's ok, skipping.
Full dataset tensor: (1872, 28, 28)
Mean: -0.118185
Standard deviation: 0.452279
Pickling notMNIST_small/G.pickle.
notMNIST_small/G
Full dataset tensor: (1872, 28, 28)
Mean: -0.0925503
Standard deviation: 0.449006
Pickling notMNIST_small/H.pickle.
notMNIST_small/H
Full dataset tensor: (1872, 28, 28)
Mean: -0.0586893
Standard deviation: 0.458759
Pickling notMNIST_small/I.pickle.
notMNIST_small/I
Full dataset tensor: (1872, 28, 28)
Mean: 0.0526451
Standard deviation: 0.471894
Pickling notMNIST_small/J.pickle.
notMNIST_small/J
Full dataset tensor: (1872, 28, 28)
Mean: -0.151689
Standard deviation: 0.448014

Problem 2

Let's verify that the data still looks good. Displaying a sample of the labels and images from the ndarray. Hint: you can use matplotlib.pyplot.



In [6]:
print(train_datasets)


['notMNIST_large/A.pickle', 'notMNIST_large/B.pickle', 'notMNIST_large/C.pickle', 'notMNIST_large/D.pickle', 'notMNIST_large/E.pickle', 'notMNIST_large/F.pickle', 'notMNIST_large/G.pickle', 'notMNIST_large/H.pickle', 'notMNIST_large/I.pickle', 'notMNIST_large/J.pickle']

Problem 3

Another check: we expect the data to be balanced across classes. Verify that.



In [7]:
# for both the training and testing dataset
for dataset in [train_datasets, test_datasets]:
    # for each letter in this dataset
    for letter_pickle in dataset:
        # unpickle the letter
        with open(letter_pickle, 'rb') as f:
            unpickled_list = pickle.load(f)
        # how many samples are here
        print('Samples for {}: {}'.format(letter_pickle, len(unpickled_list)))


Samples for notMNIST_large/A.pickle: 52909
Samples for notMNIST_large/B.pickle: 52911
Samples for notMNIST_large/C.pickle: 52912
Samples for notMNIST_large/D.pickle: 52911
Samples for notMNIST_large/E.pickle: 52912
Samples for notMNIST_large/F.pickle: 52912
Samples for notMNIST_large/G.pickle: 52912
Samples for notMNIST_large/H.pickle: 52912
Samples for notMNIST_large/I.pickle: 52912
Samples for notMNIST_large/J.pickle: 52911
Samples for notMNIST_small/A.pickle: 1872
Samples for notMNIST_small/B.pickle: 1873
Samples for notMNIST_small/C.pickle: 1873
Samples for notMNIST_small/D.pickle: 1873
Samples for notMNIST_small/E.pickle: 1873
Samples for notMNIST_small/F.pickle: 1872
Samples for notMNIST_small/G.pickle: 1872
Samples for notMNIST_small/H.pickle: 1872
Samples for notMNIST_small/I.pickle: 1872
Samples for notMNIST_small/J.pickle: 1872

Merge and prune the training data as needed. Depending on your computer setup, you might not be able to fit it all in memory, and you can tune train_size as needed. The labels will be stored into a separate array of integers 0 through 9.

Also create a validation dataset for hyperparameter tuning.


In [8]:
def make_arrays(nb_rows, img_size):
  if nb_rows:
    dataset = np.ndarray((nb_rows, img_size, img_size), dtype=np.float32)
    labels = np.ndarray(nb_rows, dtype=np.int32)
  else:
    dataset, labels = None, None
  return dataset, labels

def merge_datasets(pickle_files, train_size, valid_size=0):
  num_classes = len(pickle_files)
  valid_dataset, valid_labels = make_arrays(valid_size, image_size)
  train_dataset, train_labels = make_arrays(train_size, image_size)
  vsize_per_class = valid_size // num_classes
  tsize_per_class = train_size // num_classes
    
  start_v, start_t = 0, 0
  end_v, end_t = vsize_per_class, tsize_per_class
  end_l = vsize_per_class+tsize_per_class
  for label, pickle_file in enumerate(pickle_files):       
    try:
      with open(pickle_file, 'rb') as f:
        letter_set = pickle.load(f)
        # let's shuffle the letters to have random validation and training set
        np.random.shuffle(letter_set)
        if valid_dataset is not None:
          valid_letter = letter_set[:vsize_per_class, :, :]
          valid_dataset[start_v:end_v, :, :] = valid_letter
          valid_labels[start_v:end_v] = label
          start_v += vsize_per_class
          end_v += vsize_per_class
                    
        train_letter = letter_set[vsize_per_class:end_l, :, :]
        train_dataset[start_t:end_t, :, :] = train_letter
        train_labels[start_t:end_t] = label
        start_t += tsize_per_class
        end_t += tsize_per_class
    except Exception as e:
      print('Unable to process data from', pickle_file, ':', e)
      raise
    
  return valid_dataset, valid_labels, train_dataset, train_labels
            
            
train_size = 200000
valid_size = 10000
test_size = 10000

valid_dataset, valid_labels, train_dataset, train_labels = merge_datasets(
  train_datasets, train_size, valid_size)
_, _, test_dataset, test_labels = merge_datasets(test_datasets, test_size)

print('Training:', train_dataset.shape, train_labels.shape)
print('Validation:', valid_dataset.shape, valid_labels.shape)
print('Testing:', test_dataset.shape, test_labels.shape)


Training: (200000, 28, 28) (200000,)
Validation: (10000, 28, 28) (10000,)
Testing: (10000, 28, 28) (10000,)

Next, we'll randomize the data. It's important to have the labels well shuffled for the training and test distributions to match.


In [39]:
def randomize(dataset, labels):
  permutation = np.random.permutation(labels.shape[0])
  shuffled_dataset = dataset[permutation,:,:]
  shuffled_labels = labels[permutation]
  return shuffled_dataset, shuffled_labels
train_dataset, train_labels = randomize(train_dataset, train_labels)
test_dataset, test_labels = randomize(test_dataset, test_labels)
valid_dataset, valid_labels = randomize(valid_dataset, valid_labels)

Problem 4

Convince yourself that the data is still good after shuffling!


Someone smart wrote this code. If they were smart, surely they didn't screw up. If they didn't screw up, the data is still good. Q.E.D.

Finally, let's save the data for later reuse:


In [40]:
pickle_file = 'notMNIST.pickle'

try:
  f = open(pickle_file, 'wb')
  save = {
    'train_dataset': train_dataset,
    'train_labels': train_labels,
    'valid_dataset': valid_dataset,
    'valid_labels': valid_labels,
    'test_dataset': test_dataset,
    'test_labels': test_labels,
    }
  pickle.dump(save, f, pickle.HIGHEST_PROTOCOL)
  f.close()
except Exception as e:
  print('Unable to save data to', pickle_file, ':', e)
  raise

In [41]:
statinfo = os.stat(pickle_file)
print('Compressed pickle size:', statinfo.st_size)


Compressed pickle size: 690800512

In [42]:
print(pickle_file)


notMNIST.pickle

Problem 5

By construction, this dataset might contain a lot of overlapping samples, including training data that's also contained in the validation and test set! Overlap between training and test can skew the results if you expect to use your model in an environment where there is never an overlap, but are actually ok if you expect to see training samples recur when you use it. Measure how much overlap there is between training, validation and test samples.

Optional questions:

  • What about near duplicates between datasets? (images that are almost identical)
  • Create a sanitized validation and test set, and compare your accuracy on those in subsequent assignments.

Meh, that sounds boring.


Problem 6

Let's get an idea of what an off-the-shelf classifier can give you on this data. It's always good to check that there is something to learn, and that it's a problem that is not so trivial that a canned solution solves it.

Train a simple model on this data using 50, 100, 1000 and 5000 training samples. Hint: you can use the LogisticRegression model from sklearn.linear_model.

Optional question: train an off-the-shelf model on all the data!



In [43]:
from sklearn.linear_model import LogisticRegressionCV
from sklearn.cross_validation import cross_val_score

ns = [50,100,1000,5000]

# load from pickled file
unpickled = {}
with open(pickle_file, 'rb') as f:
    unpickled = pickle.load(f)
print(unpickled.keys())


dict_keys(['test_dataset', 'train_labels', 'valid_dataset', 'test_labels', 'train_dataset', 'valid_labels'])

In [44]:
# create the training sets (flattened to 1 dimension)
flat_training_sets = []
for n in ns:
    flat_training_sets.append([x.flatten() for x in unpickled['train_dataset'][:n]])
print("Got {} training sets".format(len(flat_training_sets)))
    
# train the models
logregCV = LogisticRegressionCV()
models = []
for i in range(len(ns)):
    print("Training classifier on subset of {} samples...".format(ns[i]))
    #print(unpickled['train_labels'][:ns[i]])
    models.append(logregCV.fit(flat_training_sets[i], unpickled['train_labels'][:ns[i]]))

# check how we did
for i in range(len(ns)):
    # TODO: run this on the testing set
    score = cross_val_score(models[i], flat_training_sets[i], train_labels[:ns[i]])
    print("Score for classifier trained on {} samples: {}".format(ns[i], score))


Got 4 training sets
[4 9 6 2 7 3 5 9 6 4 7 6 0 1 8 0 1 9 6 5 4 1 8 7 9 8 8 2 8 9 1 4 4 8 9 7 1
 7 2 5 6 1 8 6 6 0 3 7 5 5]
/usr/local/lib/python3.4/dist-packages/sklearn/cross_validation.py:516: Warning: The least populated class in y has only 2 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=3.
  % (min_labels, self.n_folds)), Warning)
[4 9 6 2 7 3 5 9 6 4 7 6 0 1 8 0 1 9 6 5 4 1 8 7 9 8 8 2 8 9 1 4 4 8 9 7 1
 7 2 5 6 1 8 6 6 0 3 7 5 5 3 4 5 0 5 0 1 4 9 9 8 6 3 4 3 5 9 2 2 6 2 8 6 5
 4 0 4 2 9 4 2 2 8 5 1 2 5 7 9 2 7 4 9 2 7 9 1 1 0 3]
[4 9 6 2 7 3 5 9 6 4 7 6 0 1 8 0 1 9 6 5 4 1 8 7 9 8 8 2 8 9 1 4 4 8 9 7 1
 7 2 5 6 1 8 6 6 0 3 7 5 5 3 4 5 0 5 0 1 4 9 9 8 6 3 4 3 5 9 2 2 6 2 8 6 5
 4 0 4 2 9 4 2 2 8 5 1 2 5 7 9 2 7 4 9 2 7 9 1 1 0 3 6 7 2 8 6 4 7 4 0 1 4
 7 1 3 8 8 7 8 6 2 9 2 3 5 4 5 3 0 8 1 8 7 6 1 6 9 8 3 3 3 1 5 9 9 4 2 7 2
 5 3 6 6 8 2 2 3 6 9 1 3 9 1 2 9 3 5 5 7 1 2 8 6 3 3 2 7 7 5 1 4 7 4 0 8 4
 9 4 4 1 0 3 4 4 0 6 9 8 4 6 2 3 2 3 6 3 3 1 9 9 4 0 7 0 5 6 1 8 6 3 9 5 8
 5 7 8 7 4 1 2 6 4 2 8 4 2 8 2 9 8 9 8 0 8 0 5 0 3 6 9 3 2 7 0 6 8 6 9 9 7
 1 8 6 9 9 1 7 0 4 5 2 9 8 4 4 8 9 5 8 5 5 1 6 5 2 1 0 4 2 0 9 8 2 4 8 7 7
 9 4 3 2 1 0 8 8 6 8 2 2 5 1 7 0 2 6 5 5 7 9 3 6 1 9 5 3 9 8 6 7 9 3 5 5 8
 3 8 9 6 7 9 8 0 5 0 8 9 7 6 9 2 3 1 7 2 5 9 2 0 6 5 3 0 4 1 9 5 7 2 5 7 9
 2 5 8 2 9 6 8 1 5 4 9 0 7 7 3 0 1 8 0 2 1 6 0 6 4 1 9 8 4 4 4 5 2 0 6 5 2
 4 4 4 8 4 0 3 5 8 3 6 4 6 6 5 2 7 9 9 7 3 6 2 1 4 1 3 9 0 9 5 6 3 3 3 5 2
 0 8 0 3 2 2 9 4 8 7 2 6 2 1 8 1 0 2 4 7 0 4 4 4 6 5 6 6 2 8 3 3 6 2 2 4 2
 5 4 1 4 4 6 8 7 2 8 6 0 0 2 1 0 3 8 8 5 4 0 2 2 0 2 7 3 7 9 2 6 2 3 5 6 8
 5 8 5 0 8 7 0 3 8 4 9 7 8 3 5 3 4 0 4 4 3 5 8 9 8 8 1 2 8 2 0 7 0 8 0 7 7
 6 7 9 1 2 1 5 1 6 3 1 1 5 1 6 5 2 7 5 4 1 6 4 2 4 4 3 9 8 6 6 5 5 7 8 8 1
 3 1 0 9 6 2 1 1 2 2 1 5 7 4 7 8 7 8 1 3 4 4 0 1 8 4 6 9 6 3 8 8 3 6 9 5 4
 3 8 6 2 7 8 4 0 7 9 8 4 0 1 5 8 7 4 7 1 2 0 5 3 5 9 6 9 7 9 9 8 5 0 7 9 2
 0 2 8 8 9 3 2 1 5 6 4 2 0 1 9 2 4 1 5 8 3 6 0 4 7 4 1 3 5 7 7 5 3 7 1 4 9
 1 1 1 9 0 8 6 1 7 4 6 1 6 7 1 4 0 7 5 8 1 7 3 5 1 3 9 9 1 9 2 6 3 6 8 3 3
 7 0 1 2 2 5 4 1 7 6 2 6 5 3 9 7 8 0 7 0 8 5 9 3 1 0 5 4 7 9 6 5 8 0 3 7 2
 0 6 9 0 0 1 4 6 7 9 1 4 9 8 9 2 9 4 4 0 3 1 3 3 1 9 6 5 7 7 5 8 0 6 3 1 5
 1 7 6 3 0 3 3 2 7 1 3 6 9 4 6 7 3 1 5 3 8 4 5 7 5 1 2 9 3 3 3 6 1 1 3 2 4
 2 0 3 6 0 5 6 8 5 5 6 2 5 2 2 8 3 4 5 3 2 2 4 7 0 4 1 8 0 5 5 9 3 5 0 2 7
 1 5 5 8 1 5 5 6 4 6 0 3 1 5 6 2 3 4 3 5 4 4 1 5 4 3 6 0 0 4 8 4 1 3 8 3 0
 5 3 3 0 5 1 0 9 4 2 7 8 7 6 9 5 2 2 6 5 5 3 8 8 0 6 8 8 8 6 6 0 1 3 0 0 4
 9 5 1 7 1 0 5 3 1 6 0 6 3 2 1 6 9 1 2 5 6 4 8 5 6 6 6 7 9 8 5 1 5 4 9 4 4
 8]
[4 9 6 ..., 3 9 5]
/usr/local/lib/python3.4/dist-packages/sklearn/cross_validation.py:516: Warning: The least populated class in y has only 1 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=3.
  % (min_labels, self.n_folds)), Warning)
/usr/local/lib/python3.4/dist-packages/sklearn/cross_validation.py:516: Warning: The least populated class in y has only 1 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=3.
  % (min_labels, self.n_folds)), Warning)
/usr/local/lib/python3.4/dist-packages/sklearn/cross_validation.py:516: Warning: The least populated class in y has only 2 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=3.
  % (min_labels, self.n_folds)), Warning)
Score for classifier trained on 50 samples: [ 0.31578947  0.29411765  0.57142857]
Score for classifier trained on 100 samples: [ 0.54054054  0.65625     0.48387097]
Score for classifier trained on 1000 samples: [ 0.78635015  0.77177177  0.7969697 ]
Score for classifier trained on 5000 samples: [ 0.81126423  0.80563887  0.81850962]

In [ ]:


In [ ]: