IndexError: index -2 is out of bounds for axis 2 with size 0

Question

I am trying to detect the defect images by using trained weight, through the error " y_pred_decoded_raw[:,:,[-2,-1]] = np.exp(y_pred_decoded_raw[:,:,[-2,-1]] * y_pred[:,:,[-2,-1]]) IndexError: index -2 is out of bounds for axis 2 with size 0".

Error is on below code

    if input_coords == 'centroids':
        y_pred_decoded_raw[:,:,[-2,-1]] = np.exp(y_pred_decoded_raw[:,:,[-2,-1]] * y_pred[:,:, 
                                          [-2,-1]]) 
        y_pred_decoded_raw[:,:,[-2,-1]] *= y_pred[:,:,[-6,-5]]
        y_pred_decoded_raw[:,:,[-4,-3]] *= y_pred[:,:,[-4,-3]] * y_pred[:,:,[-6,-5]] 
        y_pred_decoded_raw[:,:,[-4,-3]] += y_pred[:,:,[-8,-7]]
        y_pred_decoded_raw = convert_coordinates(y_pred_decoded_raw, start_index=-4, 
        conversion='centroids2corners')

Detection Module

orig_images = [] # Store the images here.
input_images = [] # Store resized versions of the images here.

batch_holder = np.zeros((30, img_height, img_width, 3))
img_dir =  "Path/to/image directory"

for i,img, in enumerate(os.listdir(img_dir)):
    img = os.path.join(img_dir, img)
    orig_images.append(imread(img))
    img =image.load_img(img, target_size=(img_height, img_width))
   
    batch_holder[i,:] =img

    y_pred = model.predict(batch_holder)
    print(y_pred)

    y_pred_decoded = decode_detections(y_pred,
                              confidence_thresh=0.5,
                              iou_threshold=0.45,
                              top_k=200,
                              input_coords='centroids',
                              normalize_coords=True,
                              img_height=img_height,
                              img_width=img_width)

    np.set_printoptions(precision=2, suppress=True, linewidth=90)
    print("Predicted boxes:\n")
    print('    class    conf  xmin     ymin   xmax    ymax')
    print(y_pred_decoded[i])

    # Display the image and draw the predicted boxes onto it.

    # Set the colors for the bounding boxes
    colors = plt.cm.hsv(np.linspace(0, 1, 21)).tolist()

    classes = ['background',
               'defect']

    plt.figure(figsize=(20, 12))
    

    current_axis = plt.gca()
    for box in y_pred_decoded[i]:
        xmin = box[-4] * orig_images[0].shape[1] / img_width
        ymin = box[-3] * orig_images[0].shape[0] / img_height
        xmax = box[-2] * orig_images[0].shape[1] / img_width
        ymax = box[-1] * orig_images[0].shape[0] / img_height
        color = colors[int(box[0])]
        label = '{}: {:.2f}'.format(classes[int(box[0])], box[1])
        current_axis.add_patch(
        plt.Rectangle((xmin, ymin), xmax - xmin, ymax - ymin, color=color, fill=False, 
        linewidth=2))
        current_axis.text(xmin, ymin, label, size='x-large', color='white', bbox={'facecolor': 
        color, 'alpha': 1.0})
      
    plt.imshow(orig_images[i])
    plt.show()