Object Detection API: Faster-RCNN, issue on Image Dimensions.

Hello everyone,

I have trained my model on Image Size: 720*700 pxs. With modification of following two files.
i. object detection / core / preprocessor.py:

def resize_image(image,
                 masks=None,
                 new_height=4160,
                 new_width=3120,
                 method=tf.image.ResizeMethod.BILINEAR,
                 align_corners=False):

ii. object detection / training / faster_rcnn_inception_v2_pets.config

model {
  faster_rcnn {
    num_classes: 10
    image_resizer {
      keep_aspect_ratio_resizer {
        min_dimension: 3120
        max_dimension: 4160
      }
    }

What is function of the both codes. If images prior training are of 600600, will these codes resize image to 31204160 ? Second, when I am using the trained model for detection and classification on Image Size: 3120*4160, images are being cropped randomly and I am not geting the whole Images in Object Detection.

Any help would be appreciated.

Best regards,
Waqar Ali Memon