Keras: Error al verificar la entrada: se esperaba que input_1 tuviera forma (299, 299, 3) pero obtuvo una matriz con forma (229, 229, 3)

Estoy tratando de entrenar un modelo inceptionv3 en keras.

Mi conjunto de datos está preprocesado en una forma 229, 229, 3.

print(data.shape)
print(type(data))
print(type(data[0]))

Salida

(1458, 229, 229, 3)

<class 'numpy.ndarray'>

<class 'numpy.ndarray'>

Inicializo mi modelo así

import os, sys
from keras.optimizers import SGD
from keras.applications import InceptionV3

model = InceptionV3()

# copile model
opt = SGD(lr=0.05)
model.compile(loss="categorical_crossentropy", optimizer=opt,
              metrics=["accuracy"])

Llamando model.fit

# train the network
print("[INFO] training network...")
H = model.fit(train_x, train_y, validation_data=(test_x, test_y),
              batch_size=batch_size, epochs=num_of_epochs, verbose=1)

Entonces me sale este error. No entiendo porque las dimensiones son correctas.

[INFO] training network...
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
 in 
      2 print("[INFO] training network...")
      3 H = model.fit(train_x, train_y, validation_data=(test_x, test_y),
----> 4               batch_size=batch_size, epochs=num_of_epochs, verbose=1)
      5 
      6 model.save(model_save_path)

~/anaconda3/lib/python3.7/site-packages/keras/engine/training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, **kwargs)
    950             sample_weight=sample_weight,
    951             class_weight=class_weight,
--> 952             batch_size=batch_size)
    953         # Prepare validation data.
    954         do_validation = False

~/anaconda3/lib/python3.7/site-packages/keras/engine/training.py in _standardize_user_data(self, x, y, sample_weight, class_weight, check_array_lengths, batch_size)
    749             feed_input_shapes,
    750             check_batch_axis=False,  # Don't enforce the batch size.
--> 751             exception_prefix='input')
    752 
    753         if y is not None:

~/anaconda3/lib/python3.7/site-packages/keras/engine/training_utils.py in standardize_input_data(data, names, shapes, check_batch_axis, exception_prefix)
    136                             ': expected ' + names[i] + ' to have shape ' +
    137                             str(shape) + ' but got array with shape ' +
--> 138                             str(data_shape))
    139     return data
    140 

ValueError: Error when checking input: expected input_1 to have shape (299, 299, 3) but got array with shape (229, 229, 3)

EDICIONES

batch_size = 32

Cómo se redimensionaron las imágenes

import imutils
import cv2

class AspectAwarePreprocessor:
    """
    CONTRUCTOR
    witdh : desired width
    height : desired height
    inter : interpolation method used when resizing the image
    """
    def __init__(self,width,height,inter=cv2.INTER_AREA):
        self.width = width
        self.height = height
        self.inter = inter

    """
    image : image to be preprocessed
    """
    def preprocess(self,image):
        # Get wdith and height of image
        (h, w) = image.shape[:2]
        dW = 0
        dH = 0

        # if width is the shorter dimension, resize image by width and crop height
        if w < h:
            image = imutils.resize(image, width=self.width,
                                   inter=self.inter)
            dH = int((image.shape[0] - self.height) / 2.0)

        # if height is the shorter dimension, resize image by height and crop width
        else:
            image = imutils.resize(image, height=self.height,
                               inter=self.inter)
            dW = int((image.shape[1] - self.width) / 2.0)

        # re-grab the width and height and use the deltas to crop the center of the image:
        (h, w) = image.shape[:2]
        image = image[dH:h - dH, dW:w - dW]

        # our image target image dimensions may be off by ± one pixel; therefore, we make a call to cv2.resize to 
        # ensure our output image has the desired width and height.
        return cv2.resize(image, (self.width, self.height),
                          interpolation=self.inter)