Question

我有以下代码：

from mnist import MNIST
from PIL import Image, ImageDraw
from sklearn.neighbors import KNeighborsClassifier
from sklearn.metrics import accuracy_score
import numpy as np
import pandas as pd
from PIL import Image, ImageFilter


def imageprepare(argv):
    """
    This function returns the pixel values.
    The imput is a png file location.
    """
    im = Image.open(argv).convert('L')
    width = float(im.size[0])
    height = float(im.size[1])
    newImage = Image.new('L', (28, 28), (255))  # creates white canvas of 28x28 pixels

    if width > height:  # check which dimension is bigger
        # Width is bigger. Width becomes 20 pixels.
        nheight = int(round((20.0 / width * height), 0))  # resize height according to ratio width
        if (nheight == 0):  # rare case but minimum is 1 pixel
            nheight = 1
            # resize and sharpen
        img = im.resize((20, nheight), Image.ANTIALIAS).filter(ImageFilter.SHARPEN)
        wtop = int(round(((28 - nheight) / 2), 0))  # calculate horizontal position
        newImage.paste(img, (4, wtop))  # paste resized image on white canvas
    else:
        # Height is bigger. Heigth becomes 20 pixels.
        nwidth = int(round((20.0 / height * width), 0))  # resize width according to ratio height
        if (nwidth == 0):  # rare case but minimum is 1 pixel
            nwidth = 1
            # resize and sharpen
        img = im.resize((nwidth, 20), Image.ANTIALIAS).filter(ImageFilter.SHARPEN)
        wleft = int(round(((28 - nwidth) / 2), 0))  # caculate vertical pozition
        newImage.paste(img, (wleft, 4))  # paste resized image on white canvas
    tv = list(newImage.getdata())  # get pixel values

    # normalize pixels to 0 and 1. 0 is pure white, 1 is pure black.
    tva = [(255 - x) * 1.0 / 255.0 for x in tv]
    print(tva)
    return tva

x=imageprepare('C:\\Users\\Wasif Ahmed\\Desktop\\9.png')#file path here
#print(len(x))# mnist IMAGES are 28x28=784 pixels


# Load dataset
mndata = MNIST('C:\\Users\\Wasif Ahmed\\Desktop\\Digit Recognition\\data\\')
images, labels = mndata.load_training()

#test_images,test_labels = mndata.load_testing()

clf = KNeighborsClassifier()

# Train on the first 10000 images:
train_x = images
train_y = labels

print("Train model")
clf.fit(train_x, train_y)

test_x = x
expected = [9]

test_x = np.array(test_x)
test_x = test_x.reshape(1, -1)

print("Compute predictions")
predicted = clf.predict(test_x)

#print("Accuracy: ", accuracy_score(expected, predicted))
print("Expected: ",expected[0])
print("predicted: ", predicted[0])

预计：9

预测：1

（我使用过https://imgur.com/a/K8P1P2p的图像）使用该功能时，imageprepare函数应该将我的图像转换为MNIST数据（我从另一篇文章中复制了它），我的预测值始终为1，好像我没有使用此函数并使用MNIST测试数据集一样，我得到了正确的预期和预测值。我怀疑我尝试将图像转换为数据的方式是错误的，因此我需要帮助才能正确执行。

test_images,test_labels = mndata.load_testing()
test_x = test_images[100]
expected = [test_labels[100]]
test_x = np.array(test_x)
test_x = test_x.reshape(1, -1)

print("Compute predictions")
predicted = clf.predict(test_x)
#print("Accuracy: ", accuracy_score(expected, predicted))
print("Expected: ",expected[0])
print("predicted: ", predicted[0])

预计：6

预计：6

如何将我自己的图像正确转换为MNIST测试数据

0 个答案: