Marcio Vinicius dos Santos mvsusp

predictor.predict should be able to handle a dictionary so tf_estimator.predict({"city":"Paris", "gender":"m", "age":22}) is the correct way to send requests like you mentioned above. The predictor logic defined here needs to handle that case.

The container source code is not open sourced yet, we are planning on releasing it very soon. I will open a bug in the container side as well, to improve the logging when the prediction fails.

I have a temporary solution to unblock you until this issue is fixed. The solution is to create a instance of a RealTimePredictor that takes a tensorflow_serving.apis request as parameter instead of a dictionary:

from sagemaker.tensorflow.predictor import tf_serializer, tf_deserializer

How to compile and debug TensorFlow

Main reference

Building TensorFlow from source

Differences

On deep learning AMI

source activate base

	#!/usr/bin/env bash

	# All the images are available under the AWS account id bellow
	ACCOUNT_ID=520713654638

	TAG=1.7.0-gpu-py2

	# get aws region
	REGION=$(aws configure get region)

	#!/usr/bin/env bash

	# All the images are available under the AWS account id bellow
	SAGEMAKER_ACCOUNT_ID=520713654638

	# The available repositories are: sagemaker-tensorflow, sagemaker-mxnet,
	# sagemaker-pytorch, sagemaker-chainer
	REPOSITORY_NAME=sagemaker-tensorflow

	aws ecr list-images --registry-id ${SAGEMAKER_ACCOUNT_ID} \




	def to_env_vars(mapping): # type: (dict) -> dict
	"""Transform a dictionary in a dictionary of env vars.
	Example:
	>>>env_vars = mapping.to_env_vars({'model_dir': '/opt/ml/model', 'batch_size': 25})
	>>>
	>>>print(args)
	['MODEL_DIR', '/opt/ml/model', 'BATCH_SIZE', 25]

	from PIL import Image
	import numpy

	image_array = numpy.random.rand(224, 224, 3) * 255

	# must of the time you wont need to generate a fake image
	image = Image.fromarray(imarray.astype('uint8')).convert('RGBA')
	image.save('result_image.png')

	class ErrorWrapper(object):
	def __init__(self, clazz):
	self.clazz = clazz

	def __missing__(self, fn):
	return getattr(self.clazz, fn, None)


	wrap_error = ErrorWrapper

	# Copyright 2018 Amazon.com, Inc. or its affiliates. All Rights Reserved.
	#
	# Licensed under the Apache License, Version 2.0 (the 'License'). You
	# may not use this file except in compliance with the License. A copy of
	# the License is located at
	#
	# http://aws.amazon.com/apache2.0/
	#
	# or in the 'license' file accompanying this file. This file is
	# distributed on an 'AS IS' BASIS, WITHOUT WARRANTIES OR CONDITIONS OF

	ChannelConfig = namedtuple_with_defaults('ChannelConfig',
	dict(ContentType="application/x-numpy", TrainingInputMode="File",
	S3DistributionType="FullyReplicated", RecordWrapperType="None"))

	from __future__ import absolute_import

	from keras import Model

	import sagemaker_container_support as scs
	from sagemaker_container_support.transformer import Transformer


	def default_model_fn(model_dir):
	return Model.load_weights(model_dir)