Getting error "Failed to find data adapter that can handle input" even after converting list to array

Question

I am getting this error : ValueError: Failed to find data adapter that can handle input' I even changed the list to arrays but still the error keeps pooping up.

This is the code:

import tensorflow.keras as keras
"""training the model starting with a random noise"""
sigma=1/30
z = make_noise("random", 32, (w, h))
losses1 = []
ms1=[]
psnr1=[]
losses = numpy.array(losses1)
ms = numpy.array(ms1)
psnr=numpy.array(psnr1)
for i in range(100):
    loss = model.fit(add_noise(z, sigma), img)
    losses.append(loss)
    y = model.predict_on_batch(z)
    m=PSNR(img1,y*255)
    print(m)
    psnr.append(m)
    if i % 20 == 0:
        y = model.predict_on_batch(z)
        image=postprocess(y[0])
        ms.append(image)
        plt.imshow(image)
        plt.show()

make_noise & add_noise functions:

"""Creating functions for necessary preprocessing and post processing"""
def preprocess(img):
    img = img_to_array(img)
    img = np.expand_dims(img, axis=0)
    img = img.astype('float32')
    img = img / 255
    return img


def postprocess(img):
    """Convert numpy array to image"""
    if len(img.shape) == 2:
        img = np.expand_dims(img, axis=-1)
    img = array_to_img(img)
    return img


def crop_image(img, d=32):
#     '''Make dimensions divisible by `d`'''

    new_size = (img.size[0] - img.size[0] % d,
                 img.size[1] - img.size[1] % d)

    bbox = [
         int((img.size[0] - new_size[0]) / 2),
         int((img.size[1] - new_size[1]) / 2),
         int((img.size[0] + new_size[0]) / 2),
         int((img.size[1] + new_size[1]) / 2),
     ]

    img_cropped = img.crop(bbox)
    return img_cropped


def get_noisy_image(img, sigma):
    """Adds Gaussian noise to an image."""
    img_noisy = np.clip(img + np.random.normal(scale=sigma, size=img.shape), 0, 1).astype(np.float32)
    return img_noisy


def make_noise(method, channel, sizes):
    """Creating a random image of range points"""
    if method == 'random':
        shape = (1, sizes[0], sizes[1], channel)
        noise = np.random.uniform(0, 0.1, size=shape)
    return noise


def add_noise(x, sigma):
    noise = np.random.normal(0, sigma, size=x.shape)
    return x + noise

```