Failed to create Input layer tensor InputPH_0

Hellow, I’m new using the Jetson TX1 and i’ve installed jetpack 3.2.1 and tensorflow 1.9 on jetson, also i’ve installed miscellaneous dependencies indicated in GitHub - NVIDIA-AI-IOT/tf_trt_models: TensorFlow models accelerated with NVIDIA TensorRT, but when i try to run the code for object detection i had a problem, this problem only occurs when i try to conver frozen model to trt model.
This the log of the problem.

trt_graph=trt.create_inference_graph(input_graph_def=frozen_graph,outputs=output_names,max_batch_size=1,max_workspace_size_bytes=1<<25,precision_mode=‘FP16’,minimum_segment_size=50)
INFO:tensorflow:Running against TensorRT version 3.0.4
INFO:tensorflow:Running against TensorRT version 3.0.4
2019-04-22 16:15:35.369286: I tensorflow/core/grappler/devices.cc:51] Number of eligible GPUs (core count >= 8): 0
2019-04-22 16:15:39.736691: I tensorflow/contrib/tensorrt/convert/convert_graph.cc:726] MULTIPLE tensorrt candidate conversion: 2
2019-04-22 16:15:39.759634: I tensorflow/contrib/tensorrt/convert/convert_nodes.cc:2390] Segment @scope ‘Postprocessor/BatchMultiClassNonMaxSuppression/MultiClassNonMaxSuppression/’, converted to graph
2019-04-22 16:15:39.809044: I tensorflow/contrib/tensorrt/convert/convert_nodes.cc:2390] Segment @scope ‘’, converted to graph
2019-04-22 16:15:39.883884: W tensorflow/contrib/tensorrt/convert/convert_graph.cc:694] Can’t determine the device, constructing an allocator at device 0
2019-04-22 16:15:39.884934: E tensorflow/contrib/tensorrt/log/trt_logger.cc:38] DefaultLogger Parameter check failed at: Network.cpp::addInput::281, condition: isIndexedCHW(dims) && volume(dims) < MAX_TENSOR_SIZE
2019-04-22 16:15:39.885062: W tensorflow/contrib/tensorrt/convert/convert_graph.cc:825] Engine creation for segment 0, composed of 181 nodes failed: Invalid argument: Failed to create Input layer tensor InputPH_0 rank=0. Skipping…
2019-04-22 16:15:39.885099: W tensorflow/contrib/tensorrt/convert/convert_graph.cc:694] Can’t determine the device, constructing an allocator at device 0

I have a doubt, jetson tx1 supports any zoo model, like faster_rcnn_inception_v2_coco, if anyone can help me, i’d be very thankful