sample_mnist.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "sample_mnist.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/desirajusantosh/DS/blob/master/sample_mnist.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "metadata": {
        "id": "NUf1tF0gH6Yy",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "%reload_ext autoreload\n",
        "%autoreload 2\n",
        "%matplotlib inline"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "0XTnwwKRH8c5",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from fastai import *\n",
        "from fastai.vision import *"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "x6m4J2ecH-i3",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "path = untar_data(URLs.MNIST_SAMPLE)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ydGh6ZV5IK16",
        "colab_type": "code",
        "outputId": "9c339c1c-d735-4f03-b91f-3acd8864ab01",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "path.ls()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[PosixPath('/root/.fastai/data/mnist_sample/train'),\n",
              " PosixPath('/root/.fastai/data/mnist_sample/labels.csv'),\n",
              " PosixPath('/root/.fastai/data/mnist_sample/valid')]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "metadata": {
        "id": "paxNvw1mIOzW",
        "colab_type": "code",
        "outputId": "f2dcc6f4-a775-4ef1-8d97-0db91576e2b1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "cell_type": "code",
      "source": [
        "(path/'train').ls()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[PosixPath('/root/.fastai/data/mnist_sample/train/7'),\n",
              " PosixPath('/root/.fastai/data/mnist_sample/train/3')]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "metadata": {
        "id": "GECB1H_lIb3f",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "tfms = get_transforms(do_flip=False)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "e5lFmkepIvpw",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "data = ImageDataBunch.from_folder(path,ds_tfms=tfms,size=26)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "0_nWwsbvI2Xk",
        "colab_type": "code",
        "outputId": "468106d2-b384-42cc-f9f0-9ba5e5af1bfe",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 491
        }
      },
      "cell_type": "code",
      "source": [
        "data.show_batch(4,figsize=(5,7))"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWYAAAHaCAYAAAAzLuMYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nbGaWmP8H5iu5og8A7QwAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 360x504 with 16 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "K-ovDyDGJAc7",
        "colab_type": "code",
        "outputId": "9f84646b-7108-48e5-cc58-dbdef5d9af4e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "data.c"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "metadata": {
        "id": "eiHgrmv0JUtt",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "77f2d5f9-782e-47f3-fa21-a659b7d14667"
      },
      "cell_type": "code",
      "source": [
        "learn = cnn_learner(data,models.resnet34)"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading: \"https://download.pytorch.org/models/resnet34-333f7ec4.pth\" to /root/.torch/models/resnet34-333f7ec4.pth\n",
            "87306240it [00:02, 32924715.09it/s]\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "metadata": {
        "id": "be-3KMXfOfT-",
        "colab_type": "code",
        "outputId": "a6809cef-f5f9-4f91-f707-28cbaa0399ba",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173
        }
      },
      "cell_type": "code",
      "source": [
        "learn.fit_one_cycle(4)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: left;\">\n",
              "      <th>epoch</th>\n",
              "      <th>train_loss</th>\n",
              "      <th>valid_loss</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>0</td>\n",
              "      <td>0.203626</td>\n",
              "      <td>0.084220</td>\n",
              "      <td>05:32</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1</td>\n",
              "      <td>0.106552</td>\n",
              "      <td>0.049734</td>\n",
              "      <td>05:21</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2</td>\n",
              "      <td>0.069282</td>\n",
              "      <td>0.024910</td>\n",
              "      <td>05:34</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3</td>\n",
              "      <td>0.062627</td>\n",
              "      <td>0.026910</td>\n",
              "      <td>05:36</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "R8ZoxgMdOkv8",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "learn.save('stage-1')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "PkGNIXxnTZ0l",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "interp = ClassificationInterpretation.from_learner(learn)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "MqxQzV9wTgh9",
        "colab_type": "code",
        "outputId": "ae030ac3-6c00-4f36-f084-92230a542cc3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 725
        }
      },
      "cell_type": "code",
      "source": [
        "interp.plot_top_losses(9, figsize=(15,11))\n"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAyAAAALECAYAAADw0lRIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/w8bBDI/rpms2gAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 1080x792 with 9 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "-lyMg1PoTjTa",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "learn.unfreeze()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "dhuzqQZXIWX4",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 80
        },
        "outputId": "28a2ea63-f59d-480a-beb0-7a62601d784d"
      },
      "cell_type": "code",
      "source": [
        "learn.fit_one_cycle(1)"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: left;\">\n",
              "      <th>epoch</th>\n",
              "      <th>train_loss</th>\n",
              "      <th>valid_loss</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>0</td>\n",
              "      <td>0.020920</td>\n",
              "      <td>0.002823</td>\n",
              "      <td>08:13</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "CGh0rUurIX-J",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "0f449d84-b54f-4f37-a9c3-6363b3946b74"
      },
      "cell_type": "code",
      "source": [
        "learn.lr_find() "
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              ""
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "kDJ7nnrNIcWu",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "outputId": "8e859875-d2d3-49d4-ed60-275b523d4396"
      },
      "cell_type": "code",
      "source": [
        "learn.recorder.plot()\n",
        "plt.title(\"Loss Vs Learning Rate\")"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Loss Vs Learning Rate')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZIAAAEWCAYAAABMoxE0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "2sFX_5x_IdXs",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 95
        },
        "outputId": "42185bde-3979-4d1b-87eb-77da9e1d248a"
      },
      "cell_type": "code",
      "source": [
        "learn.unfreeze() \n",
        "learn.fit_one_cycle(2, max_lr=slice(1e-7,1e-6))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "    <div>\n",
              "        <style>\n",
              "            /* Turns off some styling */\n",
              "            progress {\n",
              "                /* gets rid of default border in Firefox and Opera. */\n",
              "                border: none;\n",
              "                /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "                background-size: auto;\n",
              "            }\n",
              "            .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "                background: #F44336;\n",
              "            }\n",
              "        </style>\n",
              "      <progress value='0' class='' max='2', style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      0.00% [0/2 00:00<00:00]\n",
              "    </div>\n",
              "    \n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: left;\">\n",
              "      <th>epoch</th>\n",
              "      <th>train_loss</th>\n",
              "      <th>valid_loss</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "  </tbody>\n",
              "</table><p>\n",
              "\n",
              "    <div>\n",
              "        <style>\n",
              "            /* Turns off some styling */\n",
              "            progress {\n",
              "                /* gets rid of default border in Firefox and Opera. */\n",
              "                border: none;\n",
              "                /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "                background-size: auto;\n",
              "            }\n",
              "            .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "                background: #F44336;\n",
              "            }\n",
              "        </style>\n",
              "      <progress value='11' class='' max='193', style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      5.70% [11/193 00:28<07:54 0.0103]\n",
              "    </div>\n",
              "    "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "TL4wcn0nMORZ",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}