PROD_Chips.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "PROD_Chips.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "-mhx2OUeCFFy",
        "_nVTYlRzB3Z7",
        "j0uKhloSMEDx",
        "AZ_bOisXDRc5",
        "ajVQgVU4M3I_",
        "m1lzOzMCDSCo",
        "I_Dk5S5OH1eM",
        "mT0knu-YKWqC",
        "kV6wqRvSDSa8",
        "PQXy87ZQYrvi",
        "8E7fxhfAZgmn",
        "0qwpp7DsgRLH",
        "mYX45leMDSqd",
        "mLRLoXN2jgV4",
        "HkO2Ygiocd0X",
        "nm-V3qykfQLK"
      ],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/noahgift/managed_ml_systems_and_iot/blob/master/PROD_Chips.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Hat17x80OVRW",
        "colab_type": "text"
      },
      "source": [
        "# Managed Machine Learning Systems and Internet of Things\n",
        "\n",
        "![hero](https://user-images.githubusercontent.com/58792/45260894-08e67a00-b3a8-11e8-941f-e539cb10f8e1.jpg)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4h6NFsfUOaUc",
        "colab_type": "text"
      },
      "source": [
        "## Pragmatic AI Labs\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ELi2P3zcBWu-",
        "colab_type": "text"
      },
      "source": [
        "![alt text](https://paiml.com/images/logo_with_slogan_white_background.png)\n",
        "\n",
        "This notebook was produced by [Pragmatic AI Labs](https://paiml.com/).  You can continue learning about these topics by:\n",
        "\n",
        "*   Buying a copy of [Pragmatic AI: An Introduction to Cloud-Based Machine Learning](http://www.informit.com/store/pragmatic-ai-an-introduction-to-cloud-based-machine-9780134863917)\n",
        "*   Reading an online copy of [Pragmatic AI:Pragmatic AI: An Introduction to Cloud-Based Machine Learning](https://www.safaribooksonline.com/library/view/pragmatic-ai-an/9780134863924/)\n",
        "*  Watching video [Essential Machine Learning and AI with Python and Jupyter Notebook-Video-SafariOnline](https://www.safaribooksonline.com/videos/essential-machine-learning/9780135261118) on Safari Books Online.\n",
        "* Watching video [AWS Certified Machine Learning-Speciality](https://learning.oreilly.com/videos/aws-certified-machine/9780135556597)\n",
        "* Purchasing video [Essential Machine Learning and AI with Python and Jupyter Notebook- Purchase Video](http://www.informit.com/store/essential-machine-learning-and-ai-with-python-and-jupyter-9780135261095)\n",
        "*   Viewing more content at [noahgift.com](https://noahgift.com/)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-mhx2OUeCFFy",
        "colab_type": "text"
      },
      "source": [
        "# Chips\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_nVTYlRzB3Z7",
        "colab_type": "text"
      },
      "source": [
        "## Introductory Concepts in Hardware AI and IoT"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mb_rCjqF2U14",
        "colab_type": "text"
      },
      "source": [
        "### Golden Era\n",
        "\n",
        "![Golden Era of Computer Architecture](https://user-images.githubusercontent.com/58792/44999482-688cf180-af72-11e8-80ee-7c15be551bc3.png)\n",
        "\n",
        "Source: https://riscv.org/wp-content/uploads/2017/05/Mon0915-RISC-V-50-Years-Computer-Arch.pdf\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "V0Rurw6B48JV",
        "colab_type": "text"
      },
      "source": [
        "### Amhdal's Law\n",
        "\n",
        "![Amhdal's Law](https://upload.wikimedia.org/wikipedia/commons/e/ea/AmdahlsLaw.svg)\n",
        "\n",
        "\n",
        "Source:  https://upload.wikimedia.org/wikipedia/commons/e/ea/AmdahlsLaw.svg"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "j0uKhloSMEDx",
        "colab_type": "text"
      },
      "source": [
        "### ASIC vs CPU vs GPU\n",
        "\n",
        "**TPU in Production**\n",
        "![TPU Production](https://user-images.githubusercontent.com/58792/45001200-fd4a1c00-af7f-11e8-8fef-530cf2d70ada.png)\n",
        "\n",
        "**CPU**\n",
        "![How CPUs work](https://user-images.githubusercontent.com/58792/45001241-5d40c280-af80-11e8-9a82-81f900995721.png)\n",
        "\n",
        "**GPU**\n",
        "![GPUs work](https://user-images.githubusercontent.com/58792/45001287-b90b4b80-af80-11e8-9cc7-cc6d2fb2b19c.png)\n",
        "\n",
        "**TPU**\n",
        "![TPUs work](https://user-images.githubusercontent.com/58792/45001315-ef48cb00-af80-11e8-8880-aa59a272e095.png)\n",
        "\n",
        "Sources: https://storage.googleapis.com/nexttpu/index.html\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qzFAK40HCO3E",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AZ_bOisXDRc5",
        "colab_type": "text"
      },
      "source": [
        "## Emerging Hardware AI and Managed ML Platforms\n",
        "\n",
        "\n",
        "Dedicated AI Hardware + Better ML Framework Cloud-Native Abstractions\n",
        "\n",
        "\n",
        "**Examples:**\n",
        "\n",
        "* AWS Sagemaker\n",
        "* AWS DeepLense\n",
        "* Google Machine Learning Engine\n",
        "* Google BigQueryML\n",
        "* Google AutoML\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Gp2QOI5IDRpe",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ajVQgVU4M3I_",
        "colab_type": "text"
      },
      "source": [
        "## CPUs\n",
        "\n",
        "\n",
        "\n",
        "*   Stuck in 3% Yearly Performance Growth\n",
        "*   Part of the Machine Learning Solution, not **THE** solution of the future\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DzddaE_1M30f",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "m1lzOzMCDSCo",
        "colab_type": "text"
      },
      "source": [
        "## TPUs\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "x_DA-TyN09Pr",
        "colab_type": "text"
      },
      "source": [
        "TPU MNIST training code snippit\n",
        "\n",
        "```python\n",
        "python /usr/share/models/official/mnist/mnist_tpu.py \\\n",
        "  --tpu=$TPU_NAME \\\n",
        "  --data_dir=${STORAGE_BUCKET}/data \\\n",
        "  --model_dir=${STORAGE_BUCKET}/output \\\n",
        "  --use_tpu=True \\\n",
        "  --iterations=500 \\\n",
        "  --train_steps=2000\n",
        "```\n",
        "\n",
        "\n",
        "\n",
        "**Resources:**\n",
        "\n",
        "\n",
        "\n",
        "*   [Minst](https://cloud.google.com/tpu/docs/tutorials/mnist)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I_Dk5S5OH1eM",
        "colab_type": "text"
      },
      "source": [
        "## AWS Inferentia"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VgLPHfFhICoT",
        "colab_type": "text"
      },
      "source": [
        "[AWS Inferentia](https://aws.amazon.com/machine-learning/inferentia/)\n",
        "\n",
        "\n",
        "\n",
        "*   machine learning inference chip \n",
        "*   will support the TensorFlow, Apache MXNet, and PyTorch deep learning frameworks, as well as models that use the ONNX format\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mT0knu-YKWqC",
        "colab_type": "text"
      },
      "source": [
        "## Amazon Elastic Inference"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Beiajj5FK1Ol",
        "colab_type": "text"
      },
      "source": [
        "\n",
        "[Elastic Inference](https://aws.amazon.com/machine-learning/elastic-inference/)\n",
        "\n",
        "*  to attach low-cost GPU-powered acceleration to Amazon EC2 and Amazon SageMaker instances \n",
        "*  inference—can drive as much as 90% of the compute costs of the application\n",
        "*  supports TensorFlow and MXnet\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kV6wqRvSDSa8",
        "colab_type": "text"
      },
      "source": [
        "## GPUs \n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PQXy87ZQYrvi",
        "colab_type": "text"
      },
      "source": [
        "### Nvidia GPU\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "P9AunxhpnfHz",
        "colab_type": "text"
      },
      "source": [
        "\n",
        "![NVidia](https://user-images.githubusercontent.com/58792/45267130-a333d580-b41b-11e8-8548-5a773b381b8f.png)\n",
        "\n",
        "[Machine Learning with NVidia](https://www.nvidia.com/object/machine-learning.html)\n",
        "\n",
        "*  [Deep Learning Institute](https://www.nvidia.com/en-us/deep-learning-ai/education/)\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qopTUzySDSis",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8E7fxhfAZgmn",
        "colab_type": "text"
      },
      "source": [
        "### AMD GPU\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6yYvP1DGfWaK",
        "colab_type": "text"
      },
      "source": [
        "![Radeon Instinct](https://user-images.githubusercontent.com/58792/45267237-a334d500-b41d-11e8-9097-5968c53f034c.png)\n",
        "\n",
        "![Human Brain?](https://user-images.githubusercontent.com/58792/45267301-d6c42f00-b41e-11e8-8349-ca15e5f93392.png)\n",
        "\n",
        "[AMD Deep Learning](https://www.amd.com/en/graphics/servers-radeon-instinct-deep-learning)\n",
        "\n",
        "\n",
        "\n",
        "*   [Machine Learning vs Deep Learning](https://www.amd.com/en/technologies/deep-machine-learning)\n",
        "*   [ROCm](https://rocm.github.io/index.html)\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "M3VIl1ePeBK9",
        "colab_type": "code",
        "outputId": "deb6a355-2eaa-4d9c-9c80-30e769f045b2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 335
        }
      },
      "source": [
        "from IPython.display import HTML\n",
        "HTML('<iframe width=\"560\" height=\"315\" src=\"https://www.youtube.com/embed/ZSMaMKbw1y0?rel=0&amp;controls=0&amp;showinfo=0\" frameborder=\"0\" allowfullscreen></iframe>')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<iframe width=\"560\" height=\"315\" src=\"https://www.youtube.com/embed/ZSMaMKbw1y0?rel=0&amp;controls=0&amp;showinfo=0\" frameborder=\"0\" allowfullscreen></iframe>"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 1
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0qwpp7DsgRLH",
        "colab_type": "text"
      },
      "source": [
        "### Intel"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "J_cbPYdPgVYI",
        "colab_type": "text"
      },
      "source": [
        "![Intel AI Academy](https://user-images.githubusercontent.com/58792/45267399-746c2e00-b420-11e8-84de-0657e9593f78.png)\n",
        "\n",
        "![Intel Dev Cloud](https://user-images.githubusercontent.com/58792/45267467-ac27a580-b421-11e8-969f-bcca059f6536.png)\n",
        "\n",
        "[Intel AI Academy](https://software.intel.com/en-us/ai-academy)\n",
        "\n",
        "\n",
        "*   [Intel AI Academy Tensorflow Tutorial](https://software.intel.com/en-us/articles/using-tensorflow-for-deep-learning-training-and-testing)\n",
        "*   [Intel AI DevCloud](https://software.intel.com/en-us/ai-academy/devcloud)\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mYX45leMDSqd",
        "colab_type": "text"
      },
      "source": [
        "## FPGA \n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mLRLoXN2jgV4",
        "colab_type": "text"
      },
      "source": [
        "### Amazon F1 Instances\n",
        "\n",
        "![Amazon FGPA](https://user-images.githubusercontent.com/58792/45267490-3839cd00-b422-11e8-9cb9-de0c1a559280.png)\n",
        "\n",
        "![FPGA](https://user-images.githubusercontent.com/58792/45267584-06c20100-b424-11e8-892e-9fe55a84f338.png)\n",
        "\n",
        "[Amazon EC2 F1 Instances](https://aws.amazon.com/ec2/instance-types/f1/)\n",
        "\n",
        "\n",
        "\n",
        "*   [Amazon EC2 F1 Webinar](https://youtu.be/5sDVprory3c)\n",
        "*   [Reconfigure IO](https://reconfigure.io/) A platform for software defined chips\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3KaksDKBDSyD",
        "colab_type": "code",
        "outputId": "998f79bc-2149-467c-acda-96842ccc340d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 335
        }
      },
      "source": [
        "from IPython.display import HTML\n",
        "HTML('<iframe width=\"560\" height=\"315\" src=\"https://www.youtube.com/embed/5sDVprory3c?rel=0&amp;controls=0&amp;showinfo=0\" frameborder=\"0\" allowfullscreen></iframe>')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<iframe width=\"560\" height=\"315\" src=\"https://www.youtube.com/embed/5sDVprory3c?rel=0&amp;controls=0&amp;showinfo=0\" frameborder=\"0\" allowfullscreen></iframe>"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HkO2Ygiocd0X",
        "colab_type": "text"
      },
      "source": [
        "## A11/12 Chip\n",
        "\n",
        "Apple Designed Chip for iOS that includes integrated GPU and dedicated neural network hardware."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PgbVUbVKtz1A",
        "colab_type": "text"
      },
      "source": [
        "[A11 \"Neural Engine\" Chip from Apple](https://en.wikipedia.org/wiki/Apple_A11)\n",
        "![A11](https://user-images.githubusercontent.com/58792/45267788-6e7a4b00-b428-11e8-99c2-e80827961480.jpg)\n",
        "![CreateML](https://user-images.githubusercontent.com/58792/45268030-03cb0e80-b42c-11e8-9a47-b8eef57e74a6.png)\n",
        "\n",
        "\n",
        "*  [600 billion operations per second](https://www.apple.com/newsroom/2017/09/the-future-is-here-iphone-x/)\n",
        "*  [Metal for Accelerating Machine Learning](https://developer.apple.com/videos/play/wwdc2018/609/)\n",
        "*  [Introducing Create ML](https://developer.apple.com/videos/play/wwdc2018/703/)\n",
        "*  [What's New in Core ML](https://developer.apple.com/videos/play/wwdc2018/708/)\n",
        "*  [AWS Lambda + Swift](https://medium.com/capital-one-developers/serverless-computing-with-swift-f515ff052919) Tangentially related...running Swift code in AWS Lambda\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Oaz3fzwvcd0Z",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6JRSfoXScqc-",
        "colab_type": "text"
      },
      "source": [
        "## AWS IoT Ecosystem\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ct4ISjzyLxWQ",
        "colab_type": "text"
      },
      "source": [
        "Many New Services from ***Reinvent 2018***\n",
        "\n",
        "* [IoT Graph](https://aws.amazon.com/about-aws/whats-new/2018/11/iot-things-graph-now-in-preview/)\n",
        "* [IoT Events](https://aws.amazon.com/about-aws/whats-new/2018/11/introducing-aws-iot-events-now-available-in-preview/)\n",
        "* [IoT SiteWise](https://aws.amazon.com/about-aws/whats-new/2018/11/announcing-aws-iot-sitewise-now-available-in-limited-preview/)\n",
        "* [IoT Device Tester](https://aws.amazon.com/about-aws/whats-new/2018/11/bluetooth-low-energy-ble-support-in-amazon-freertos-now-available-in-beta/)\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QD75VVPNyGLe",
        "colab_type": "text"
      },
      "source": [
        "![DeepLense ](https://user-images.githubusercontent.com/58792/45337586-a6bf7d80-b53d-11e8-9796-ecef716335a8.png)\n",
        "\n",
        "Greengrass part of Embedded Machine Intelligence Platform\n",
        "\n",
        "[AWS Greengrass](https://aws.amazon.com/greengrass/)\n",
        "\n",
        "**Dev Kits**\n",
        "\n",
        "![aws iot dev kit](https://user-images.githubusercontent.com/58792/46483952-a6697980-c7ad-11e8-84e5-90beb17a460f.jpg)\n",
        "\n",
        "![AWS IoT](https://user-images.githubusercontent.com/58792/46484897-8935aa80-c7af-11e8-9f73-940d1dca9ad1.jpg)\n",
        "\n",
        "**Resource Examples:**\n",
        "\n",
        "* [DeepLens AWS Blog](https://aws.amazon.com/blogs/aws/deeplens/)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nm-V3qykfQLK",
        "colab_type": "text"
      },
      "source": [
        "## Azure IoT Ecosystem"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nRbbApUMn9b6",
        "colab_type": "text"
      },
      "source": [
        "[Azure IoT kits](https://microsoft.github.io/azure-iot-developer-kit/)\n",
        "\n",
        "Integrated into Visual Studio via Azure IoT Workbench\n",
        "\n",
        "![mxchip](https://user-images.githubusercontent.com/58792/46483506-a9179f00-c7ac-11e8-8c48-3a87f49c12f9.jpg)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dI9IwAyfcqc_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ju6xqhsJw-MR",
        "colab_type": "text"
      },
      "source": [
        "## GPU Demos"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Re9jdU5nxBYx",
        "colab_type": "text"
      },
      "source": [
        "### Mandelbrot"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RizeFcTRRSBP",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 211
        },
        "outputId": "02ab9502-596c-4600-9f30-19bd3a7ad5c4"
      },
      "source": [
        "%matplotlib inline\n",
        "from matplotlib import pyplot as plt\n",
        "import numpy as np\n",
        "import math\n",
        "from numba import jit, njit, vectorize, cuda"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: \n",
            "Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so.\n",
            "\n",
            "For more information about alternatives visit: ('http://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup')\n",
            "  warnings.warn(errors.NumbaWarning(msg))\n",
            "/usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: \n",
            "Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda-10.0/nvvm/libdevice.\n",
            "\n",
            "For more information about alternatives visit: ('http://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup')\n",
            "  warnings.warn(errors.NumbaWarning(msg))\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_9_HoXwERH1i",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 156
        },
        "outputId": "d5af34db-393d-43ee-e473-4d170d9893ed"
      },
      "source": [
        "!pip install numba\n",
        "!find / -iname 'libdevice'\n",
        "!find / -iname 'libnvvm.so'"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: numba in /usr/local/lib/python3.6/dist-packages (0.47.0)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from numba) (1.17.5)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from numba) (45.1.0)\n",
            "Requirement already satisfied: llvmlite>=0.31.0dev0 in /usr/local/lib/python3.6/dist-packages (from numba) (0.31.0)\n",
            "/usr/local/cuda-10.0/nvvm/libdevice\n",
            "/usr/local/cuda-10.1/nvvm/libdevice\n",
            "/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so\n",
            "/usr/local/cuda-10.1/nvvm/lib64/libnvvm.so\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jZEihqKCREZx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import os\n",
        "os.environ['NUMBAPRO_LIBDEVICE'] = \"/usr/local/cuda-10.0/nvvm/libdevice\"\n",
        "os.environ['NUMBAPRO_NVVM'] = \"/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so\""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9iISjADi2D_Y",
        "colab_type": "text"
      },
      "source": [
        "##### Non-GPU Mandelbrot"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "APrhGfCC1xlG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "from pylab import imshow, show\n",
        "from timeit import default_timer as timer"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T6KviMLo10I9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def mandel(x, y, max_iters):\n",
        "  \"\"\"\n",
        "    Given the real and imaginary parts of a complex number,\n",
        "    determine if it is a candidate for membership in the Mandelbrot\n",
        "    set given a fixed number of iterations.\n",
        "  \"\"\"\n",
        "  c = complex(x, y)\n",
        "  z = 0.0j\n",
        "  for i in range(max_iters):\n",
        "    z = z*z + c\n",
        "    if (z.real*z.real + z.imag*z.imag) >= 4:\n",
        "      return i\n",
        "\n",
        "  return max_iters"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_2K9mG5P12b1",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def create_fractal(min_x, max_x, min_y, max_y, image, iters):\n",
        "  height = image.shape[0]\n",
        "  width = image.shape[1]\n",
        "\n",
        "  pixel_size_x = (max_x - min_x) / width\n",
        "  pixel_size_y = (max_y - min_y) / height\n",
        "    \n",
        "  for x in range(width):\n",
        "    real = min_x + x * pixel_size_x\n",
        "    for y in range(height):\n",
        "      imag = min_y + y * pixel_size_y\n",
        "      color = mandel(real, imag, iters)\n",
        "      image[y, x] = color"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3SdUxzp-19Jc",
        "colab_type": "code",
        "outputId": "dd8a7a8c-ea0c-464c-f9b9-e306f81bf6ad",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        }
      },
      "source": [
        "image = np.zeros((1024, 1536), dtype = np.uint8)\n",
        "start = timer()\n",
        "create_fractal(-2.0, 1.0, -1.0, 1.0, image, 20) \n",
        "dt = timer() - start\n",
        "\n",
        "print(\"Mandelbrot created in %f s\" % dt)\n",
        "imshow(image)\n",
        "show()"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mandelbrot created in 4.369222 s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nqhAzfVMAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3eBtMqwD2I4f",
        "colab_type": "text"
      },
      "source": [
        "##### Numba Mandelbrot"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Jjzp4kAj7yN5",
        "colab_type": "text"
      },
      "source": [
        "Numba is [LLVM JIT](http://numba.pydata.org/) support for Python"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "P1M03OI67s08",
        "colab_type": "text"
      },
      "source": [
        "###### Run Numba Mandelbrot"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d8IlJnNu2e1W",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from numba import jit\n",
        "\n",
        "@jit\n",
        "def mandel(x, y, max_iters):\n",
        "  \"\"\"\n",
        "    Given the real and imaginary parts of a complex number,\n",
        "    determine if it is a candidate for membership in the Mandelbrot\n",
        "    set given a fixed number of iterations.\n",
        "  \"\"\"\n",
        "  c = complex(x, y)\n",
        "  z = 0.0j\n",
        "  for i in range(max_iters):\n",
        "    z = z*z + c\n",
        "    if (z.real*z.real + z.imag*z.imag) >= 4:\n",
        "      return i\n",
        "\n",
        "  return max_iters\n",
        "\n",
        "@jit\n",
        "def create_fractal(min_x, max_x, min_y, max_y, image, iters):\n",
        "  height = image.shape[0]\n",
        "  width = image.shape[1]\n",
        "\n",
        "  pixel_size_x = (max_x - min_x) / width\n",
        "  pixel_size_y = (max_y - min_y) / height\n",
        "    \n",
        "  for x in range(width):\n",
        "    real = min_x + x * pixel_size_x\n",
        "    for y in range(height):\n",
        "      imag = min_y + y * pixel_size_y\n",
        "      color = mandel(real, imag, iters)\n",
        "      image[y, x] = color"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "g0ovBdjl2rsn",
        "colab_type": "code",
        "outputId": "08855a5d-caaf-4bc9-9549-f1b4cff171ff",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 366
        }
      },
      "source": [
        "image = np.zeros((1024, 1536), dtype = np.uint8)\n",
        "start = timer()\n",
        "create_fractal(-2.0, 1.0, -1.0, 1.0, image, 20) \n",
        "dt = timer() - start\n",
        "\n",
        "print(\"Mandelbrot created in %f s\" % dt)\n",
        "imshow(image)\n",
        "show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mandelbrot created in 0.310116 s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeoAAAFMCAYAAAAEKP/JAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f0d3e663c88>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "13WPkJ5P2ud1",
        "colab_type": "text"
      },
      "source": [
        "##### CUDA GPU Mandelbrot"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Bg00L6HA21Oe",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from numba import cuda\n",
        "from numba import *\n",
        "\n",
        "mandel_gpu = cuda.jit(restype=uint32, argtypes=[f8, f8, uint32], device=True)(mandel)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "M31bUz5Y22sx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "@cuda.jit(argtypes=[f8, f8, f8, f8, uint8[:,:], uint32])\n",
        "def mandel_kernel(min_x, max_x, min_y, max_y, image, iters):\n",
        "  height = image.shape[0]\n",
        "  width = image.shape[1]\n",
        "\n",
        "  pixel_size_x = (max_x - min_x) / width\n",
        "  pixel_size_y = (max_y - min_y) / height\n",
        "\n",
        "  startX, startY = cuda.grid(2)\n",
        "  gridX = cuda.gridDim.x * cuda.blockDim.x;\n",
        "  gridY = cuda.gridDim.y * cuda.blockDim.y;\n",
        "\n",
        "  for x in range(startX, width, gridX):\n",
        "    real = min_x + x * pixel_size_x\n",
        "    for y in range(startY, height, gridY):\n",
        "      imag = min_y + y * pixel_size_y \n",
        "      image[y, x] = mandel_gpu(real, imag, iters)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hfH-xwR_7WX0",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        },
        "outputId": "744eadec-74a5-4451-8044-b8708951f4ab"
      },
      "source": [
        "image = np.zeros((1024, 1536), dtype = np.uint8)\n",
        "start = timer()\n",
        "create_fractal(-2.0, 1.0, -1.0, 1.0, image, 20) \n",
        "dt = timer() - start\n",
        "\n",
        "print(\"Mandelbrot created in %f s\" % dt)\n",
        "imshow(image)\n",
        "show()"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mandelbrot created in 0.344193 s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nqhAzfVMAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9yeFvfri9qC6",
        "colab_type": "text"
      },
      "source": [
        "##### Cuda Vectorize"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "apd8ic7m-BZB",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 225
        },
        "outputId": "6d7c8e4d-b04d-4c62-c9a8-1821216590ed"
      },
      "source": [
        "from numba import (cuda, vectorize)\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from sklearn.cluster import KMeans\n",
        "\n",
        "from functools import wraps\n",
        "from time import time\n",
        "\n",
        "def real_estate_df():\n",
        "    \"\"\"30 Years of Housing Prices\"\"\"\n",
        "\n",
        "    df = pd.read_csv(\"https://raw.githubusercontent.com/noahgift/real_estate_ml/master/data/Zip_Zhvi_SingleFamilyResidence.csv\")\n",
        "    df.rename(columns={\"RegionName\":\"ZipCode\"}, inplace=True)\n",
        "    df[\"ZipCode\"]=df[\"ZipCode\"].map(lambda x: \"{:.0f}\".format(x))\n",
        "    df[\"RegionID\"]=df[\"RegionID\"].map(lambda x: \"{:.0f}\".format(x))\n",
        "    return df\n",
        "\n",
        "def numerical_real_estate_array(df):\n",
        "    \"\"\"Converts df to numpy numerical array\"\"\"\n",
        "\n",
        "    columns_to_drop = ['RegionID', 'ZipCode', 'City', 'State', 'Metro', 'CountyName']\n",
        "    df_numerical = df.dropna()\n",
        "    df_numerical = df_numerical.drop(columns_to_drop, axis=1)\n",
        "    return df_numerical.values\n",
        "\n",
        "def real_estate_array():\n",
        "    \"\"\"Returns Real Estate Array\"\"\"\n",
        "\n",
        "    df = real_estate_df()\n",
        "    rea = numerical_real_estate_array(df)\n",
        "    return np.float32(rea)\n",
        "\n",
        "\n",
        "@vectorize(['float32(float32, float32)'], target='cuda')\n",
        "def add_ufunc(x, y):\n",
        "    return x + y\n",
        "\n",
        "def cuda_operation():\n",
        "    \"\"\"Performs Vectorized Operations on GPU\"\"\"\n",
        "\n",
        "    x = real_estate_array()\n",
        "    y = real_estate_array()\n",
        "\n",
        "    print(\"Moving calculations to GPU memory\")\n",
        "    x_device = cuda.to_device(x)\n",
        "    y_device = cuda.to_device(y)\n",
        "    out_device = cuda.device_array(\n",
        "        shape=(x_device.shape[0],x_device.shape[1]), dtype=np.float32)\n",
        "    print(x_device)\n",
        "    print(x_device.shape)\n",
        "    print(x_device.dtype)\n",
        "\n",
        "    print(\"Calculating on GPU\")\n",
        "    add_ufunc(x_device,y_device, out=out_device)\n",
        "\n",
        "    out_host = out_device.copy_to_host()\n",
        "    print(f\"Calculations from GPU {out_host}\")\n",
        "\n",
        "cuda_operation()"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Moving calculations to GPU memory\n",
            "<numba.cuda.cudadrv.devicearray.DeviceNDArray object at 0x7fd32514df60>\n",
            "(10015, 259)\n",
            "float32\n",
            "Calculating on GPU\n",
            "Calculations from GPU [[2.0000e+00 8.4160e+05 8.4700e+05 ... 2.1086e+06 2.0970e+06 2.0888e+06]\n",
            " [4.0000e+00 1.0848e+06 1.0934e+06 ... 3.0316e+06 3.0398e+06 3.0506e+06]\n",
            " [6.0000e+00 1.4180e+05 1.4240e+05 ... 2.2760e+05 2.2700e+05 2.2660e+05]\n",
            " ...\n",
            " [3.0554e+04 1.9780e+05 1.9620e+05 ... 4.3960e+05 4.4000e+05 4.4080e+05]\n",
            " [3.0560e+04 1.7340e+05 1.7500e+05 ... 3.8140e+05 3.8460e+05 3.8720e+05]\n",
            " [3.0562e+04 5.0700e+05 5.0800e+05 ... 1.5672e+06 1.5590e+06 1.5484e+06]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t_SAHKhgd_AJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "omSTDQwaCftQ",
        "colab_type": "text"
      },
      "source": [
        "# References\n",
        "\n",
        "\n",
        "*   [Pragmatic AI: An Introduction to Cloud-Based Machine Learning-Physical Book](http://www.informit.com/store/pragmatic-ai-an-introduction-to-cloud-based-machine-9780134863917)\n",
        "*   [Pragmatic AI: Pragmatic AI: An Introduction to Cloud-Based Machine Learning-SafariOnline Book](https://www.safaribooksonline.com/library/view/pragmatic-ai-an/9780134863924/)\n",
        "*  [Essential Machine Learning and AI with Python and Jupyter Notebook-Video-SafariOnline](https://www.safaribooksonline.com/videos/essential-machine-learning/9780135261118)\n",
        "* [Essential Machine Learning and AI with Python and Jupyter Notebook- Purchase Video](http://www.informit.com/store/essential-machine-learning-and-ai-with-python-and-jupyter-9780135261095)\n",
        "*  [noahgift.com](https://noahgift.com/)\n",
        "*  [paiml.com](https://paiml.com/)\n",
        "*  https://storage.googleapis.com/nexttpu/index.html\n",
        "* https://riscv.org/wp-content/uploads/2017/05/Mon0915-RISC-V-50-Years-Computer-Arch.pdf\n",
        "* https://software.intel.com/en-us/ai-academy/devcloud\n",
        "* [David Patterson Says It’s Time for New Computer Architectures and Software Languages](https://spectrum.ieee.org/view-from-the-valley/computing/hardware/david-patterson-says-its-time-for-new-computer-architectures-and-software-languages)\n",
        "* [Nuclear Powered Command Line Tools](https://github.com/noahgift/nuclear_powered_command_line_tools)\n",
        "* [CUDA Colab Install](https://medium.com/@iphoenix179/running-cuda-c-c-in-jupyter-or-how-to-run-nvcc-in-google-colab-663d33f53772)\n",
        "* [Forbes: How Vertically Integrated AI Stacks Will Affect IT Organizations](https://www.forbes.com/sites/forbestechcouncil/2018/11/02/how-vertically-integrated-ai-stacks-will-affect-it-organizations/#7a2f6f691713)\n",
        "* [Forbes:  Cloud Native Machine Learning and AI](https://www.forbes.com/sites/forbestechcouncil/2018/07/05/cloud-native-machine-learning-and-ai/#34c136ab5397)\n",
        "*   [Google Machine Learning Crash Course](https://developers.google.com/machine-learning/crash-course/)\n",
        "\n",
        "\n"
      ]
    }
  ]
}