You must feed a value for placeholder tensor 'import/input_tensor' with dtype string and shape

[PapaMadeleine2022]

first, I download resnet_v2_fp32_savedmodel_NHWC_jpg saved_model from blog

then, I run a example provided in nvidia documents about trt-tf here :

# Import TensorFlow and TensorRT
import tensorflow as tf
import tensorflow.contrib.tensorrt as trt
# Inference with TF-TRT `SavedModel` workflow:
graph = tf.Graph()
with graph.as_default():
    with tf.Session() as sess:
        # Create a TensorRT inference graph from a SavedModel:
        trt_graph = trt.create_inference_graph(
            input_graph_def=None,
            outputs=None,
            input_saved_model_dir="/tmp/resnet/1538687457",
            input_saved_model_tags=["serve"]) 
        # Import the TensorRT graph into a new graph and run:
        output_node = tf.import_graph_def(
            trt_graph,
            return_elements=["softmax_tensor"])
        sess.run(output_node)

but it shows error:

2019-03-19 17:21:53.332397: I tensorflow/contrib/tensorrt/convert/convert_graph.cc:913] Number of TensorRT candidate segments: 1
2019-03-19 17:21:54.040844: W tensorflow/contrib/tensorrt/convert/convert_nodes.cc:3710] Validation failed for TensorRTInputPH_0 and input slot 0: Input tensor with shape [?,?,?,3] has an unknown non-batch dimension at dim 1
2019-03-19 17:21:54.040922: W tensorflow/contrib/tensorrt/convert/convert_graph.cc:1021] TensorRT node TRTEngineOp_0 added for segment 0 consisting of 445 nodes failed: Invalid argument: Validation failed for TensorRTInputPH_0 and input slot 0: Input tensor with shape [?,?,?,3] has an unknown non-batch dimension at dim 1. Fallback to TF...
2019-03-19 17:21:54.289076: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:581] Optimization results for grappler item: tf_graph
2019-03-19 17:21:54.289142: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:583]   constant folding: Graph size after: 550 nodes (-256), 613 edges (-258), time = 820.515ms.
2019-03-19 17:21:54.289160: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:583]   layout: Graph size after: 557 nodes (7), 615 edges (2), time = 142.796ms.
2019-03-19 17:21:54.289171: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:583]   constant folding: Graph size after: 552 nodes (-5), 615 edges (0), time = 535.989ms.
2019-03-19 17:21:54.289184: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:583]   TensorRTOptimizer: Graph size after: 552 nodes (0), 615 edges (0), time = 1296.37695ms.
2019-03-19 17:22:03.134267: I tensorflow/stream_executor/dso_loader.cc:152] successfully opened CUDA library libcublas.so.10.0 locally
Traceback (most recent call last):
  File "trt-tf.py", line 18, in <module>
    sess.run(output_node)
  File "/xxx/.virtualenvs/tf-tmp/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 929, in run
    run_metadata_ptr)
  File "/xxx/.virtualenvs/tf-tmp/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1152, in _run
    feed_dict_tensor, options, run_metadata)
  File "/xxx/.virtualenvs/tf-tmp/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1328, in _do_run
    run_metadata)
  File "/xxx/.virtualenvs/tf-tmp/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1348, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: You must feed a value for placeholder tensor 'import/input_tensor' with dtype string and shape [?]
	 [[node import/input_tensor (defined at trt-tf.py:17) ]]
	 [[node import/map/Shape (defined at trt-tf.py:17) ]]

Caused by op u'import/input_tensor', defined at:
  File "trt-tf.py", line 17, in <module>
    return_elements=["softmax_tensor"])
  File "/xxx/.virtualenvs/tf-tmp/lib/python2.7/site-packages/tensorflow/python/util/deprecation.py", line 507, in new_func
    return func(*args, **kwargs)
  File "/xxx/.virtualenvs/tf-tmp/lib/python2.7/site-packages/tensorflow/python/framework/importer.py", line 442, in import_graph_def
    _ProcessNewOps(graph)
  File "/xxx/.virtualenvs/tf-tmp/lib/python2.7/site-packages/tensorflow/python/framework/importer.py", line 235, in _ProcessNewOps
    for new_op in graph._add_new_tf_operations(compute_devices=False):  # pylint: disable=protected-access
  File "/xxx/.virtualenvs/tf-tmp/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 3433, in _add_new_tf_operations
    for c_op in c_api_util.new_tf_operations(self)
  File "/xxx/.virtualenvs/tf-tmp/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 3325, in _create_op_from_tf_operation
    ret = Operation(c_op, self)
  File "/xxx/.virtualenvs/tf-tmp/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 1801, in __init__
    self._traceback = tf_stack.extract_stack()

InvalidArgumentError (see above for traceback): You must feed a value for placeholder tensor 'import/input_tensor' with dtype string and shape [?]
	 [[node import/input_tensor (defined at trt-tf.py:17) ]]
	 [[node import/map/Shape (defined at trt-tf.py:17) ]]

how to fix it? anyone can give some advises?

my envs:
tensorflow-gpu 1.13.1

[eduardinjo]

@IvyGongoogle did you solve the issue? I'm having similar problem