Tensorflow object detection, Parser error: BoxPredictor_0/Reshape: Reshape: -1 dimension specified m...

My environment is:
GPU:Telsa M40
Tensorflow:1.8
tf object detection version: up-to-data version clone from git(clone at yesterday)
od extractor :“ssd_mobilenet_v1”
input size: 800x800
my config.py is as follows:

import graphsurgeon as gs
import tensorflow as tf

Input = gs.create_node("Input",
    op="Placeholder",
    dtype=tf.float32,
    shape=[1, 3, 800, 800])
PriorBox = gs.create_node("PriorBox",
    numLayers=6,
    minScale=0.03,
    maxScale=0.80,
    aspectRatios=[1.0, 2.0, 0.5, 3.0, 0.33],
    layerVariances=[0.1,0.1,0.2,0.2],
    featureMapShapes=[50, 25, 13, 7, 4, 2])
   NMS = gs.create_node("NMS",
    scoreThreshold=1e-8,
    iouThreshold=0.6,
    maxDetectionsPerClass=100,
    maxTotalDetections=100,
    numClasses=8,
    scoreConverter="SIGMOID")
concat_priorbox = gs.create_node("concat_priorbox", dtype=tf.float32, axis=2)
concat_box_loc = gs.create_node("concat_box_loc")
concat_box_conf = gs.create_node("concat_box_conf")

namespace_plugin_map = {
    "MultipleGridAnchorGenerator": PriorBox,
    "Postprocessor": NMS,
    "Preprocessor": Input,
     "ToFloat": Input,
    "image_tensor": Input,
    "MultipleGridAnchorGenerator/Concatenate": concat_priorbox,
    "concat": concat_box_loc,
    "concat_1": concat_box_conf
}

def preprocess(dynamic_graph):
    # Now create a new graph by collapsing namespaces
    dynamic_graph.collapse_namespaces(namespace_plugin_map)
    # Remove the outputs, so we just have a single output node (NMS).

by the way, what version of tf detection model fit tensorRT 4.0.
thank you a lot.

Hello, what error are you seeing? The error in the subject is too long and cut off.