im new in Deep Learning ( actually i´m taking some courses ). I tried to do binary classification example with a CNN and i´m not able to predict. I want to check ear corn color images ( 20 x 28 x 3 )

My code:

import tensorflow as tf
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import datetime
from tensorflow.keras.layers import Input, Conv2D, Dense, Flatten, Dropout, GlobalMaxPooling2D,BatchNormalization
from tensorflow.keras.models import Model
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from PIL import Image
import cv2
import glob
from pathlib import Path
import os

image_path = 'sample_data/ears'
images = [cv2.imread(file) for file in glob.glob("sample_data/ears/*.png")]

def append_images(images):
  appendedImages = []
  for image in images:
    img=np.array(image)
    resized_image = cv2.resize(img, (28, 20))
    print(resized_image.shape)
    appendedImages.append(resized_image)
  return appendedImages

x_train = append_images(images)

for i, img in enumerate(x_train):
    img_pil = Image.fromarray(img)
    img_pil.save(os.path.join("sample_data/resized_ears", f'corn_kernel_{i}.png'))

print(f'Se guardaron {len(x_train)} imágenes en {"sample_data/resized_ears"}')

data_dir = 'sample_data/ears/resized_ears'

datagen = ImageDataGenerator(rescale=1./255, validation_split=0.2)

train_generator = datagen.flow_from_directory(
    data_dir,
    batch_size=32,
    target_size=(28, 20),
    class_mode='binary',
    subset='training'
    )

validation_generator = datagen.flow_from_directory(
    data_dir,
    target_size=(28, 20),
    batch_size=32,
    class_mode='binary',
    subset='validation'
  )

print("Classes found:", train_generator.class_indices)
print("Classes found:", validation_generator.class_indices)

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense

model = Sequential([
    Conv2D(32, (3, 3), activation='relu', input_shape=(20, 28, 3)),
    MaxPooling2D((2, 2)),
    Conv2D(64, (3, 3), activation='relu'),
    MaxPooling2D((2, 2)),
    Flatten(),
    Dense(64, activation='relu'),
    Dense(1, activation='sigmoid')
])


model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])

model.fit(train_generator, epochs=10)


images = [cv2.imread(file) for file in glob.glob("sample_data/ears/resized_ears/class1/*.png")]

images[0].shape

model.predict(images[0])

The model does the fit

Epoch 1/10
1/1 [==============================] - 1s 1s/step - loss: 0.7027 - accuracy: 0.0000e+00
Epoch 2/10
1/1 [==============================] - 0s 30ms/step - loss: 0.3947 - accuracy: 1.0000
Epoch 3/10
1/1 [==============================] - 0s 30ms/step - loss: 0.2136 - accuracy: 1.0000
Epoch 4/10
1/1 [==============================] - 0s 31ms/step - loss: 0.1110 - accuracy: 1.0000
Epoch 5/10
1/1 [==============================] - 0s 33ms/step - loss: 0.0505 - accuracy: 1.0000
Epoch 6/10
1/1 [==============================] - 0s 32ms/step - loss: 0.0206 - accuracy: 1.0000
Epoch 7/10
1/1 [==============================] - 0s 30ms/step - loss: 0.0078 - accuracy: 1.0000
Epoch 8/10
1/1 [==============================] - 0s 30ms/step - loss: 0.0029 - accuracy: 1.0000
Epoch 9/10
1/1 [==============================] - 0s 32ms/step - loss: 0.0010 - accuracy: 1.0000
Epoch 10/10
1/1 [==============================] - 0s 30ms/step - loss: 3.8640e-04 - accuracy: 1.0000
<keras.src.callbacks.History at 0x7e1b08663b20>

But when I try to do the prediction i got this error message:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-157-0f33f27dad40> in <cell line: 5>()
      3 images[0].shape
      4 
----> 5 model.predict(images[0])
      6 

1 frames
/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py in tf__predict_function(iterator)
     13                 try:
     14                     do_return = True
---> 15                     retval_ = ag__.converted_call(ag__.ld(step_function), (ag__.ld(self), ag__.ld(iterator)), None, fscope)
     16                 except:
     17                     do_return = False

ValueError: in user code:

    File "/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py", line 2440, in predict_function  *
        return step_function(self, iterator)
    File "/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py", line 2425, in step_function  **
        outputs = model.distribute_strategy.run(run_step, args=(data,))
    File "/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py", line 2413, in run_step  **
        outputs = model.predict_step(data)
    File "/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py", line 2381, in predict_step
        return self(x, training=False)
    File "/usr/local/lib/python3.10/dist-packages/keras/src/utils/traceback_utils.py", line 70, in error_handler
        raise e.with_traceback(filtered_tb) from None
    File "/usr/local/lib/python3.10/dist-packages/keras/src/engine/input_spec.py", line 298, in assert_input_compatibility
        raise ValueError(

    ValueError: Input 0 of layer "sequential_23" is incompatible with the layer: expected shape=(None, 20, 28, 3), found shape=(None, 28, 3)

Why i´m having this?.