Predicting the class of image with CNN

Question

I am trying to predict the class of single image on trained model, but I am getting a strange output, so this is my code:

from tensorflow.keras.models import load_model
from tensorflow.keras.preprocessing import image
import matplotlib.pyplot as plt
import numpy as np
import os


def load_image(img_path, show=False):
  img = image.load_img(img_path, target_size=(300, 300))
  img_tensor = image.img_to_array(img)                    
  img_tensor = np.expand_dims(img_tensor, axis=0)         
  img_tensor /= 255.                                      

  if show:
    plt.imshow(img_tensor[0])
    plt.axis('off')
    plt.show()

  return img_tensor


if __name__ == "__main__":
  # load model
  model = load_model("model1.h5")

  # image path
  img_path = 'dog.jpg' # dog

  # load a single image
  new_image = load_image(img_path, True)

  # check prediction
  pred = model.predict(new_image)
  print(pred)

But I am getting [[0.8189566 0.18104333]] as output. But I have classes 0 and 1. Could this be because the batch size is not specified anywhere?

Answer 1

with model.predict(new_image) you obtain the probability of each test image to belong to a particular class

to get the output class, you need to select the class with the max probability predicted:

np.argmax(pred, axis=1)