io19-demo-1-swift-in-colab.ipynb

io19-demo-1-swift-in-colab.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "io19-demo-1-swift-in-colab.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "swift",
      "display_name": "Swift"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/rakesh4real/1d8238ac665c491a233c8a822f9970d6/io19-demo-1-swift-in-colab.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xSYmKIt59Zt3",
        "colab_type": "text"
      },
      "source": [
        "**Credits:** Google I/O 19 "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "g8sVjqcc-ej8",
        "colab_type": "text"
      },
      "source": [
        "# Hello, Colab!"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kZRlD4utdPuX",
        "colab_type": "code",
        "outputId": "8a0b2b2f-85a9-4208-b450-25c939121e34",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "let pi = 3.14\n",
        "let y = [1, 2, 3]\n",
        "print(\"Hello\")"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Hello\r\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ym3klNgB9iA4",
        "colab_type": "code",
        "outputId": "b41187af-e83b-4064-cb98-33c8e2802706",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "source": [
        "let someNumbers = [\n",
        "        \"Prime\": [2, 3, 5, 7, 11, 13],\n",
        "    \"Fibonacci\": [1, 1, 2, 3, 5, 8],\n",
        "       \"Square\": [1, 4, 9, 16, 25],\n",
        "]\n",
        "\n",
        "for (kind, numbers) in someNumbers {\n",
        "    var largest = 0\n",
        "    for x in numbers where x > largest && x.isEven {\n",
        "        largest = x\n",
        "    }\n",
        "    print(\"The largest even \\(kind) is \\(largest)\")\n",
        "}"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The largest even Fibonacci is 8\r\n",
            "The largest even Square is 16\r\n",
            "The largest even Prime is 2\r\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "igwEhJod9qla",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "extension Int{\n",
        "  var isEven: Bool{\n",
        "    return isMultiple(of: 2)\n",
        "  }\n",
        "}"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LVbNpcWY-gea",
        "colab_type": "text"
      },
      "source": [
        "# The Tensor Library"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CYKFHSzl9yUS",
        "colab_type": "code",
        "outputId": "9d44ed40-6ec8-4f20-e3f3-a09997652975",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 119
        }
      },
      "source": [
        "\n",
        "import TensorFlow\n",
        "\n",
        "let x: Tensor<Float> = [[[1, 2], [4, 5]]]\n",
        "\n",
        "print(\"x + x =\\n\\(x + x)\\n\")\n",
        "print(\"sum = \\(x.sum())\")\n",
        "print(\"stdev = \\(x.standardDeviation(squeezingAxes: 0, 1))\")"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "x + x =\r\n",
            "[[[ 2.0,  4.0],\r\n",
            "  [ 8.0, 10.0]]]\r\n",
            "\r\n",
            "sum = 12.0\r\n",
            "stdev = [1.5, 1.5]\r\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xo-oB7g1-vct",
        "colab_type": "text"
      },
      "source": [
        "# Deep Learning"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vV_giCKJ-noW",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "struct Model: Layer {\n",
        "    var conv = Conv2D<Float>(filterShape: (5, 5, 3, 6))\n",
        "    var maxpool = MaxPool2D<Float>(poolSize: (2, 2), strides: (2, 2))\n",
        "    var flatten = Flatten<Float>()\n",
        "    var dense = Dense<Float>(inputSize: 36 * 6, outputSize: 10)\n",
        "\n",
        "    @differentiable\n",
        "    func callAsFunction(_ input: Tensor<Float>) -> Tensor<Float> {\n",
        "        return input.sequenced(through: conv, maxpool, flatten, dense )\n",
        "    }\n",
        "}"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DAi4uvSV-1lq",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "// Use random training data.\n",
        "let x = Tensor<Float>(randomNormal: [10, 16, 16, 3])\n",
        "let y = Tensor<Int32>(rangeFrom: 0, to: 10, stride: 1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GfMUZTNN_oV6",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "var model = Model()\n",
        "let optimizer = SGD(for: model)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "j5gVGbNQ_tda",
        "colab_type": "code",
        "outputId": "ebc0133b-09f7-4916-a2e7-0dc7cfaa9f37",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 187
        }
      },
      "source": [
        "for i in 1...10 {\n",
        "    let (loss, grads) = valueWithGradient(at: model) { model -> Tensor<Float> in\n",
        "        let logits = model(x)\n",
        "        return softmaxCrossEntropy(logits: logits, labels: y)\n",
        "    }\n",
        "    print(\"Step \\(i), loss is: \\(loss)\")\n",
        "    optimizer.update(&model.allDifferentiableVariables, along: grads)\n",
        "}"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Step 1, loss is: 3.16299\r\n",
            "Step 2, loss is: 2.758711\r\n",
            "Step 3, loss is: 2.4353342\r\n",
            "Step 4, loss is: 2.1696808\r\n",
            "Step 5, loss is: 1.9441502\r\n",
            "Step 6, loss is: 1.7495903\r\n",
            "Step 7, loss is: 1.5788343\r\n",
            "Step 8, loss is: 1.4273479\r\n",
            "Step 9, loss is: 1.2923636\r\n",
            "Step 10, loss is: 1.1733923\r\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}