Allow local mode and add image

Author: Dan Choi

advanced_functionality/tensorflow_bring_your_own/data/cat.png

@@ -213,20 +213,36 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "# https://docs.docker.com/compose/gettingstarted/#step-2-create-a-dockerfile\n",
-      "FROM tensorflow/tensorflow:1.8.0-py3\n",
-      "\n",
-      "RUN apt-get update && apt-get install -y --no-install-recommends nginx curl\n",
-      "\n",
-      "# Download tensorflow serving\n",
-      "RUN echo \"deb [arch=amd64] http://storage.googleapis.com/tensorflow-serving-apt stable tensorflow-model-server tensorflow-model-server-universal\" | tee /etc/apt/sources.list.d/tensorflow-serving.list\n",
-      "RUN curl https://storage.googleapis.com/tensorflow-serving-apt/tensorflow-serving.release.pub.gpg | apt-key add -\n",
-      "RUN apt-get update && apt-get install tensorflow-model-server\n",
-      "\n",
-      "ENV PATH=\"/opt/ml/code:${PATH}\"\n",
-      "\n",
-      "COPY /cifar10 /opt/ml/code\n",
-      "WORKDIR /opt/ml/code\n"
+      "# Copyright 2017-2018 Amazon.com, Inc. or its affiliates. All Rights Reserved.\r\n",
+      "#\r\n",
+      "# Licensed under the Apache License, Version 2.0 (the \"License\"). You\r\n",
+      "# may not use this file except in compliance with the License. A copy of\r\n",
+      "# the License is located at\r\n",
+      "#\r\n",
+      "#     http://aws.amazon.com/apache2.0/\r\n",
+      "#\r\n",
+      "# or in the \"license\" file accompanying this file. This file is\r\n",
+      "# distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF\r\n",
+      "# ANY KIND, either express or implied. See the License for the specific\r\n",
+      "# language governing permissions and limitations under the License.\r\n",
+      "\r\n",
+      "# For more information on creating a Dockerfile\r\n",
+      "# https://docs.docker.com/compose/gettingstarted/#step-2-create-a-dockerfile\r\n",
+      "FROM tensorflow/tensorflow:1.8.0-py3\r\n",
+      "\r\n",
+      "RUN apt-get update && apt-get install -y --no-install-recommends nginx curl\r\n",
+      "\r\n",
+      "# Download TensorFlow Serving\r\n",
+      "# https://www.tensorflow.org/serving/setup#installing_the_modelserver\r\n",
+      "RUN echo \"deb [arch=amd64] http://storage.googleapis.com/tensorflow-serving-apt stable tensorflow-model-server tensorflow-model-server-universal\" | tee /etc/apt/sources.list.d/tensorflow-serving.list\r\n",
+      "RUN curl https://storage.googleapis.com/tensorflow-serving-apt/tensorflow-serving.release.pub.gpg | apt-key add -\r\n",
+      "RUN apt-get update && apt-get install tensorflow-model-server\r\n",
+      "\r\n",
+      "ENV PATH=\"/opt/ml/code:${PATH}\"\r\n",
+      "\r\n",
+      "# /opt/ml and all subdirectories are utilized by SageMaker, we use the /code subdirectory to store our user code.\r\n",
+      "COPY /cifar10 /opt/ml/code\r\n",
+      "WORKDIR /opt/ml/code"
      ]
     }
    ],
@@ -310,48 +326,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "/anaconda3/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
-      "  from ._conv import register_converters as _register_converters\n",
-      "Download from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz and extract.\n",
-      "WARNING:tensorflow:From utils/generate_cifar10_tfrecords.py:45: maybe_download (from tensorflow.contrib.learn.python.learn.datasets.base) is deprecated and will be removed in a future version.\n",
-      "Instructions for updating:\n",
-      "Please write your own downloading logic.\n",
-      "WARNING:tensorflow:From /anaconda3/lib/python3.6/site-packages/tensorflow/contrib/learn/python/learn/datasets/base.py:252: _internal_retry.<locals>.wrap.<locals>.wrapped_fn (from tensorflow.contrib.learn.python.learn.datasets.base) is deprecated and will be removed in a future version.\n",
-      "Instructions for updating:\n",
-      "Please use urllib or similar directly.\n",
-      "Successfully downloaded cifar-10-python.tar.gz 170498071 bytes.\n",
-      "Generating /tmp/cifar-10-data/train.tfrecords\n",
-      "Generating /tmp/cifar-10-data/validation.tfrecords\n",
-      "Generating /tmp/cifar-10-data/eval.tfrecords\n",
-      "Removing original files.\n",
-      "Done!\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "! python utils/generate_cifar10_tfrecords.py --data-dir=/tmp/cifar-10-data"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "eval.tfrecords       train.tfrecords      validation.tfrecords\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "! ls /tmp/cifar-10-data "
    ]
@@ -373,7 +359,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -401,71 +387,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO:sagemaker:Creating training-job with name: tensorflow_cifar10_example-2018-08-01-08-13-48-683\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Attaching to tmp1mcb1n6k_algo-1-ASVN5_1\n",
-      "Creating tmp1mcb1n6k_algo-1-ASVN5_1 ... \n",
-      "\u001b[36malgo-1-ASVN5_1  |\u001b[0m Training complete.\n",
-      "\u001b[1B\u001b[36mtmp1mcb1n6k_algo-1-ASVN5_1 exited with code 0\n",
-      "\u001b[0mAborting on container exit...\n",
-      "===== Job Complete =====\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO:sagemaker:Creating model with name: tensorflow_cifar10_example-2018-08-01-08-14-21-042\n",
-      "INFO:sagemaker:Creating endpoint with name tensorflow_cifar10_example-2018-08-01-08-13-48-683\n",
-      "WARNING:urllib3.connectionpool:Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPConnection object at 0x10ba8d5c0>: Failed to establish a new connection: [Errno 61] Connection refused',)': /ping\n",
-      "WARNING:urllib3.connectionpool:Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPConnection object at 0x10ba8d048>: Failed to establish a new connection: [Errno 61] Connection refused',)': /ping\n",
-      "WARNING:urllib3.connectionpool:Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPConnection object at 0x10bc12748>: Failed to establish a new connection: [Errno 61] Connection refused',)': /ping\n",
-      "WARNING:urllib3.connectionpool:Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'ProtocolError('Connection aborted.', RemoteDisconnected('Remote end closed connection without response',))': /ping\n",
-      "WARNING:urllib3.connectionpool:Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'ProtocolError('Connection aborted.', RemoteDisconnected('Remote end closed connection without response',))': /ping\n",
-      "WARNING:urllib3.connectionpool:Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'ProtocolError('Connection aborted.', RemoteDisconnected('Remote end closed connection without response',))': /ping\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Attaching to tmpi66tdkz1_algo-1-DZB6N_1\n",
-      "Creating tmpi66tdkz1_algo-1-DZB6N_1 ... \n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m Starting the inference server with 2 workers.\n",
-      "\u001b[1B\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:22.729603: I tensorflow_serving/model_servers/main.cc:153] Building single TensorFlow model file config:  model_name: cifar10_model model_base_path: /opt/ml/model/export/Servo\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:22.730876: I tensorflow_serving/model_servers/server_core.cc:459] Adding/updating models.\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:22.730991: I tensorflow_serving/model_servers/server_core.cc:514]  (Re-)adding model: cifar10_model\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:22.850950: I tensorflow_serving/core/basic_manager.cc:716] Successfully reserved resources to load servable {name: cifar10_model version: 1533111259}\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:22.851019: I tensorflow_serving/core/loader_harness.cc:66] Approving load for servable version {name: cifar10_model version: 1533111259}\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:22.851048: I tensorflow_serving/core/loader_harness.cc:74] Loading servable version {name: cifar10_model version: 1533111259}\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:22.852366: I external/org_tensorflow/tensorflow/contrib/session_bundle/bundle_shim.cc:360] Attempting to load native SavedModelBundle in bundle-shim from: /opt/ml/model/export/Servo/1533111259\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:22.853667: I external/org_tensorflow/tensorflow/cc/saved_model/loader.cc:242] Loading SavedModel with tags: { serve }; from: /opt/ml/model/export/Servo/1533111259\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:22.872867: I external/org_tensorflow/tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:22.918331: I external/org_tensorflow/tensorflow/cc/saved_model/loader.cc:161] Restoring SavedModel bundle.\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:22.982049: I external/org_tensorflow/tensorflow/cc/saved_model/loader.cc:196] Running LegacyInitOp on SavedModel bundle.\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:23.016319: I external/org_tensorflow/tensorflow/cc/saved_model/loader.cc:291] SavedModel load for tags { serve }; Status: success. Took 163280 microseconds.\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:23.018543: I tensorflow_serving/servables/tensorflow/saved_model_warmup.cc:83] No warmup data file found at /opt/ml/model/export/Servo/1533111259/assets.extra/tf_serving_warmup_requests\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:23.033650: I tensorflow_serving/core/loader_harness.cc:86] Successfully loaded servable version {name: cifar10_model version: 1533111259}\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:23.038377: I tensorflow_serving/model_servers/main.cc:323] Running ModelServer at 0.0.0.0:8500 ...\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m [warn] getaddrinfo: address family for nodename not supported\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 2018-08-01 08:14:23.041668: I tensorflow_serving/model_servers/main.cc:333] Exporting HTTP/REST API at:localhost:8501 ...\n",
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m [evhttp_server.cc : 235] RAW: Entering the event loop ...\n",
-      "!\u001b[36malgo-1-DZB6N_1  |\u001b[0m 172.18.0.1 - - [01/Aug/2018:08:14:23 +0000] \"GET /ping HTTP/1.1\" 200 2 \"-\" \"-\"\n"
-     ]
-    }
-   ],
+   "outputs": [],
+   "source": [
+    "!/bin/bash ./utils/setup.sh"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
     "from sagemaker.estimator import Estimator\n",
     "\n",
@@ -490,49 +423,46 @@
    "source": [
     "## Making predictions using Python SDK\n",
     "\n",
-    "In order to do some predictions, we will use an image, which will be converted using OpenCV, into a json format to send as an inference request."
+    "In order to do some predictions, we will use an image, which will be converted using OpenCV, into a json format to send as an inference request. We need to install OpenCV as we using that to deserialize our image that we will use for predictions."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[36malgo-1-DZB6N_1  |\u001b[0m 172.18.0.1 - - [01/Aug/2018:08:14:28 +0000] \"POST /invocations HTTP/1.1\" 200 236 \"-\" \"-\"\n"
+      "Collecting opencv-python\n",
+      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/53/e0/21c8964fa8ef50842ebefaa7346a3cf0e37b56c8ecd97ed6bd2dbe577705/opencv_python-3.4.2.17-cp36-cp36m-manylinux1_x86_64.whl (25.0MB)\n",
+      "\u001b[K    100% |████████████████████████████████| 25.0MB 2.1MB/s eta 0:00:01\n",
+      "\u001b[?25hRequirement already satisfied: numpy>=1.11.3 in /home/ec2-user/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages (from opencv-python) (1.14.5)\n",
+      "\u001b[31mdistributed 1.21.8 requires msgpack, which is not installed.\u001b[0m\n",
+      "Installing collected packages: opencv-python\n",
+      "Successfully installed opencv-python-3.4.2.17\n",
+      "\u001b[33mYou are using pip version 10.0.1, however version 18.0 is available.\n",
+      "You should consider upgrading via the 'pip install --upgrade pip' command.\u001b[0m\n"
      ]
-    },
-    {
-     "data": {
-      "text/plain": [
-       "{'predictions': [{'probabilities': [0.0203898,\n",
-       "    0.00186337,\n",
-       "    0.00125851,\n",
-       "    0.116194,\n",
-       "    0.860205,\n",
-       "    2.24463e-10,\n",
-       "    3.87913e-06,\n",
-       "    8.29227e-05,\n",
-       "    1.4244e-06,\n",
-       "    1.60669e-06],\n",
-       "   'classes': 4}]}"
-      ]
-     },
-     "execution_count": 4,
-     "metadata": {},
-     "output_type": "execute_result"
     }
    ],
+   "source": [
+    "! pip install opencv-python"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
     "import cv2\n",
     "import numpy\n",
     "\n",
     "from sagemaker.predictor import json_serializer, json_deserializer\n",
     "\n",
-    "image = cv2.imread(\"/path/to/image.png\", 1)\n",
+    "image = cv2.imread(\"data/cat.png\", 1)\n",
     "\n",
     "# resize, as our model is expecting images in 32x32.\n",
     "image = cv2.resize(image, (32, 32))\n",
@@ -554,22 +484,24 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "INFO:sagemaker:Deleting endpoint with name: tensorflow_cifar10_example-2018-08-01-08-13-48-683\n"
+      "INFO:sagemaker:Deleting endpoint with name: tensorflow_cifar10_example-2018-08-03-18-06-55-168\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
       "Gracefully stopping... (press Ctrl+C again to force)\n",
-      "Stopping tmpi66tdkz1_algo-1-DZB6N_1 ... \n",
+      "Stopping tmp3n0u5hj2_algo-1-HCRIC_1 ... \r\n",
+      "\u001b[1A\u001b[2K\r",
+      "Stopping tmp3n0u5hj2_algo-1-HCRIC_1 ... \u001b[32mdone\u001b[0m\r",
       "\u001b[1B"
      ]
     }
@@ -591,7 +523,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -610,7 +542,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -632,7 +564,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -648,7 +580,7 @@
     "## Training on SageMaker\n",
     "Training on SageMaker with the Python SDK is done in the same fashion as training locally. This is done by changing our train_instance_type from `local` to one of our [supported EC2 instance types](https://aws.amazon.com/sagemaker/pricing/instance-types/).\n",
     "\n",
-    "In addition, we must now specify an ECR image url, which we just pushed above.\n",
+    "In addition, we must now specify an ECR image url, which we just pushed above. Please make sure to replace the string within the Estimator parameter, image_name.\n",
     "\n",
     "Finally, our local training dataset has to be in Amazon S3 and the S3 url to our dataset will be passed into the `fit()` call."
    ]
@@ -703,7 +635,7 @@
     }
    ],
    "source": [
-    "image = cv2.imread(\"/path/to/image.png\", 1)\n",
+    "image = cv2.imread(\"data/cat.png\", 1)\n",
     "\n",
     "# resize, as our model is expecting images in 32x32.\n",
     "image = cv2.resize(image, (32, 32))\n",
@@ -789,9 +721,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "conda_tensorflow_p36",
    "language": "python",
-   "name": "python3"
+   "name": "conda_tensorflow_p36"
   },
   "language_info": {
    "codemirror_mode": {
@@ -803,7 +735,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.5"
+   "version": "3.6.4"
   }
  },
  "nbformat": 4,

advanced_functionality/tensorflow_bring_your_own/tensorflow_bring_your_own.ipynb

@@ -0,0 +1,10 @@
+
+{
+	"default-runtime": "nvidia",
+    "runtimes": {
+        "nvidia": {
+            "path": "/usr/bin/nvidia-container-runtime",
+            "runtimeArgs": []
+        }
+    }
+}