Principal Component Analysis not working

Question

I'm trying to do principal component analysis on datasets containing images, but whenever I want to apply pca.transform from the sklearn.decomposition module I keep getting this error: *AttributeError: 'PCA' object has no attribute 'mean_'*. I know what this error means, but I have no clue how to fix it. I reckon some of you guys know how to fix this.

Thank you for your help

My code:

from sklearn import svm
import numpy as np
import glob
import os
from PIL import Image
from sklearn.decomposition import PCA

image_dir1 = "C:\Users\private\Desktop\K FOLDER\private\train"
image_dir2 = "C:\Users\private\Desktop\K FOLDER\private\test1"
Standard_size = (300,200)
pca = PCA(n_components = 10)
file_open = lambda x,y: glob.glob(os.path.join(x,y))


def matrix_image(image_path):
    "opens image and converts it to a m*n matrix" 
    image = Image.open(image_path)
    print("changing size from %s to %s" % (str(image.size), str(Standard_size)))
    image = image.resize(Standard_size)
    image = list(image.getdata())
    image = map(list,image)
    image = np.array(image)
    return image
def flatten_image(image):  
    """
    takes in a n*m numpy array and flattens it to 
    an array of the size (1,m*n)
    """
    s = image.shape[0] * image.shape[1]
    image_wide = image.reshape(1,s)
    return image_wide[0]

if __name__ == "__main__":
    train_images = file_open(image_dir1,"*.jpg")
    test_images = file_open(image_dir2,"*.jpg")
    train_set = []
    test_set = []

    "Loop over all images in files and modify them"
    train_set = [flatten_image(matrix_image(image)) for image in train_images]
    test_set = [flatten_image(matrix_image(image)) for image in test_images]
    train_set = np.array(train_set)
    test_set = np.array(test_set)
    train_set = pca.fit_transform(train_set) "line where error occurs"
    test_set = pca.fit_transform(test_set)

Full traceback:

Traceback (most recent call last):
  File "C:\Users\Private\workspace\final_submission\src\d.py", line 54, in <module>
    train_set = pca.transform(train_set)
  File "C:\Python27\lib\site-packages\sklearn\decomposition\pca.py", line 298, in transform
    if self.mean_ is not None:
AttributeError: 'PCA' object has no attribute 'mean_'

Edit1: So I tried to fit the model before transforming it, and now I'm getting an even weirder error. I looked it up, and it involves f2py, a module that ports Fortran to Python which is part of the Numpy Library.

File "C:\Users\Private\workspace\final_submission\src\d.py", line 54, in <module>
    pca.fit(train_set)
  File "C:\Python27\lib\site-packages\sklearn\decomposition\pca.py", line 200, in fit
    self._fit(X)
  File "C:\Python27\lib\site-packages\sklearn\decomposition\pca.py", line 249, in _fit
    U, S, V = linalg.svd(X, full_matrices=False)
  File "C:\Python27\lib\site-packages\scipy\linalg\decomp_svd.py", line 100, in svd
    full_matrices=full_matrices, overwrite_a = overwrite_a)
ValueError: failed to create intent(cache|hide)|optional array-- must have defined dimensions but got (0,)

Edit2:

So I have checked if my train_set and data_set contained any data and they don't. I've checked my image_dirs, and they contain the right locations(just for clarity, I got them by going to the actual files, looking at the properties of one the images and copied the location). The fault should lie somewhere else.

Answer 1

You should fit the model before transform:

train_set = np.array(train_set)
test_set = np.array(test_set)

pca.fit(train_set)
pca.fit(test_set)

train_set = pca.transform(train_set) "line where error occurs"
test_set = pca.transform(test_set)

Edit

Second error indicate that your train_set is empty. It can be easily reproduced using this code:

train_set = np.array([[]])
pca.fit(train_set)

I think one problem is in flatten_image function. I may be wrong but this line will raise AttributeError

image.wide = image.reshape(1,s)

It can be replaced with:

image_wide = image.reshape(1,s)
return image_wide[0]

This line is problematic too:

print("changing size from %s to %s" % str(image.size), str(Standard_size))

Read http://docs.python.org/2/library/stdtypes.html#string-formatting-operations for more details, but values must be a tuple. So you want this instead:

print("changing size from %s to %s" % (str(image.size), str(Standard_size)))

Another edit

At last you replace loops aftert "Loop over all images in files and modify them" with:

train_set = [flatten_image(matrix_image(image)) for image in train_images]
test_set = [flatten_image(matrix_image(image)) for image in test_images]

Right now you call file_open so it will look for files in path like this: "C:\Users\private\Desktop\K FOLDER\private\train\C:\Users\private\Desktop\K FOLDER\private\train\foo.jpg" and you get empty list instead of file name.