How to fix problem 'Error when checking input: expected flatten_7_input to have shape (28, 28) but got array with shape (28, 3)'

Question

I'm trying to create the NN using example on Tensorflow and feed it my own hand-written digit to predict right label but shape of array doesn't allow me to do this.

import os
import cv2
import numpy as np
import matplotlib.pyplot as plt
import tensorflow as tf
mnist = tf.keras.datasets.mnist
(x_train, y_train),(x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0

model = tf.keras.models.Sequential([
  tf.keras.layers.Flatten(input_shape=(28, 28)),
  tf.keras.layers.Dense(512, activation=tf.nn.relu),
  tf.keras.layers.Dropout(0.2),
  tf.keras.layers.Dense(10, activation=tf.nn.softmax)
    ])
model.compile(optimizer='adam',
      loss='sparse_categorical_crossentropy',
      metrics=['accuracy'])

model.fit(x_train, y_train, epochs=5)
model.evaluate(x_test, y_test)

path = 'C:/Users/pewdu/Desktop/third.jpg'
img = cv2.imread(path)
new_img = cv2.resize(img, (28, 28))
new_img = new_img / 255.0
print(new_img.shape) # it equals to (28,28,3)
prediction = model.predict(new_img)

So the error is:

ValueError: Error when checking input: expected flatten_7_input to have shape (28, 28) but got array with shape (28, 3)

Answer 1

The Error message says it all.

The model you initialized expects data in a N x W x H format where,

• N = Number of examples
• W = Width of the image
• H = Height of the image

When you are reading the image using cv2.imread() , you can see that the image size is given in W x H x C format where,

• W = Width of image
• H = Height of image
• C = Number of channels in the image (3 for RGB/BGR and 1 for GrayScale)

Your model is expecting a grayscale normalized image. And since you are sending only one image (single example), you need to reshape your image array by adding an axis at the front.

img = cv2.imread(path)
new_img = cv2.resize(img, (28, 28))
new_img = new_img[:,:,0] / 255.0 # Take only first channel and normalize
new_img = np.expand_dims(new_img, axis=0) # Adding the dimension
print(new_img.shape) # it equals to (1, 28, 28)
prediction = model.predict(new_img)