fd2.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Untitled11.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPoru/pesEQ7CNZHljx0PWX",
      "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/rohanreddych/symmetrical-guacamole/blob/master/dib/fd2.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-6f1xHBhIzOx",
        "colab_type": "code",
        "outputId": "62a80e55-03c6-4ce9-fc92-fbc2aae6e2e5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        }
      },
      "source": [
        "%tensorflow_version 2.x"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "TensorFlow 2.x selected.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uQPKylV53rET",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import cv2\n",
        "import os\n",
        "import numpy as np"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LmmzcliM4RrU",
        "colab_type": "code",
        "outputId": "f1580b5a-092d-42b6-a751-c7616d302f10",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        }
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CGmn7uBuWd2l",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow.keras import layers\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "M8y2Cip5Wgl5",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "f=open(\"/content/drive/My Drive/names.txt\",'r')\n",
        "#f = open('', \"r\")\n",
        "\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L7h0Z5dPX4P-",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "names = f.readlines()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0p2HyqeZX7Mr",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def rstrip(x):\n",
        "    x = x.rstrip(\" \\n\")\n",
        "    return x"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t-iQRee5X_v0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "names = list(map(rstrip, names))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Lq21kdM_YOFC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "bin_dict = \"/content/drive/My Drive/full_data/bin/\"\n",
        "gray_dict = \"/content/drive/My Drive/full_data/gray/\""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dHehFZyrYVsy",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "X = [gray_dict + i for i in names]\n",
        "y = [bin_dict + i for i in names]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t2uyAgY6YZqv",
        "colab_type": "code",
        "outputId": "63fce6c0-d5dd-426f-e779-465db9b0daef",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        }
      },
      "source": [
        "print(len(X), len(y))\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "43264 43264\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xytFPuyWYbbA",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def pp(img, lbl):\n",
        "    img = tf.io.read_file(img)\n",
        "    img = tf.io.decode_png(img, channels=1)\n",
        "    img = tf.image.resize(img, [104,104])\n",
        "    img = img / 255.\n",
        "    lbl = tf.io.read_file(lbl)\n",
        "    lbl = tf.image.decode_png(lbl, channels=1)\n",
        "    lbl = tf.image.resize(lbl, [104,104])\n",
        "    lbl = lbl / 255.\n",
        "    lbl = tf.round(lbl)\n",
        "    return img, lbl\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "E6eOBB1UYcT3",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "ds = tf.data.Dataset.from_tensor_slices((X,y))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EMwLXLpZXyAE",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model = tf.keras.models.Sequential([layers.InputLayer(input_shape=(104,104,1)),\n",
        "                                    layers.Conv2D(64, (3,3), activation=\"relu\", padding=\"same\"),\n",
        "                                    layers.MaxPool2D((2,2), padding=\"same\"),\n",
        "                                    layers.Conv2D(32, (3,3), activation=\"relu\", padding=\"same\"),\n",
        "                                    layers.MaxPool2D((2,2), padding=\"same\"),\n",
        "                                    layers.Conv2D(16, (3,3), activation=\"relu\", padding=\"same\"),\n",
        "                                    layers.MaxPool2D((2,2), padding=\"same\"),\n",
        "                                    layers.Conv2D(16, (3,3), activation=\"relu\", padding=\"same\"),\n",
        "                                    layers.UpSampling2D((2,2)),\n",
        "                                    layers.Conv2D(32, (3,3), activation=\"relu\", padding=\"same\"),\n",
        "                                    layers.UpSampling2D((2,2)),\n",
        "                                    layers.Conv2D(64, (3,3), activation=\"relu\", padding=\"same\"),\n",
        "                                    layers.UpSampling2D((2,2)),\n",
        "                                    layers.Conv2D(1, (3,3), padding=\"same\"),\n",
        "                                    \n",
        "])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MM42xyAXYJ8h",
        "colab_type": "code",
        "outputId": "4b6f6862-8a7f-4a10-d1f2-ae1cbbe71722",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 663
        }
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d (Conv2D)              (None, 104, 104, 64)      640       \n",
            "_________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D) (None, 52, 52, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_1 (Conv2D)            (None, 52, 52, 32)        18464     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 26, 26, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 26, 26, 16)        4624      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2 (None, 13, 13, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_3 (Conv2D)            (None, 13, 13, 16)        2320      \n",
            "_________________________________________________________________\n",
            "up_sampling2d (UpSampling2D) (None, 26, 26, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_4 (Conv2D)            (None, 26, 26, 32)        4640      \n",
            "_________________________________________________________________\n",
            "up_sampling2d_1 (UpSampling2 (None, 52, 52, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_5 (Conv2D)            (None, 52, 52, 64)        18496     \n",
            "_________________________________________________________________\n",
            "up_sampling2d_2 (UpSampling2 (None, 104, 104, 64)      0         \n",
            "_________________________________________________________________\n",
            "conv2d_6 (Conv2D)            (None, 104, 104, 1)       577       \n",
            "=================================================================\n",
            "Total params: 49,761\n",
            "Trainable params: 49,761\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lzqcATrkYLqY",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model.compile(loss=\"mse\", optimizer=\"adam\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TbD5QSulCC3c",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "ds2 = ds.map(pp)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kGAyJe06dTU0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "gray_l = []\n",
        "bin_l = []\n",
        "for i , j in ds2.take(100):\n",
        "    gray_l.append(i)\n",
        "    bin_l.append(j)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4et580uxg4tV",
        "colab_type": "code",
        "outputId": "ec1df1e2-9545-4667-cf0d-b9369d9af5b1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        }
      },
      "source": [
        "plt.imshow(bin_l[69].numpy().reshape(104,104), cmap=\"gray\")\n",
        "#print(bin_l[4].numpy())"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f6806151898>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 53
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQMAAAD7CAYAAABuZ/ELAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJwMzS5wMzAxwMjCzxMnAzAAnAzNL/g/bedOcU18wrAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QUk0YZa-kUnr",
        "colab_type": "code",
        "outputId": "551bff3f-1bb4-4d01-e32e-e4aa89721bd8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        }
      },
      "source": [
        "plt.imshow(gray_l[69].numpy().reshape(104,104), cmap=\"gray\")"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f67ecf7cba8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 54
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQMAAAD7CAYAAABuZ/ELAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bRSJGPsQkgAp",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "ds3 = ds2.batch(32)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ONjzlwt3P7H1",
        "colab_type": "code",
        "outputId": "02c2f162-2d72-41fc-9e8c-8c38aeb7036a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 399
        }
      },
      "source": [
        "model.fit(ds3, epochs=10)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train for 1352 steps\n",
            "Epoch 1/10\n",
            "1352/1352 [==============================] - 14406s 11s/step - loss: 0.0390\n",
            "Epoch 2/10\n",
            "1352/1352 [==============================] - 397s 293ms/step - loss: 0.0222\n",
            "Epoch 3/10\n",
            "1352/1352 [==============================] - 396s 293ms/step - loss: 0.0192\n",
            "Epoch 4/10\n",
            "1352/1352 [==============================] - 398s 295ms/step - loss: 0.0178\n",
            "Epoch 5/10\n",
            "1352/1352 [==============================] - 400s 296ms/step - loss: 0.0168\n",
            "Epoch 6/10\n",
            "1352/1352 [==============================] - 404s 299ms/step - loss: 0.0161\n",
            "Epoch 7/10\n",
            "1352/1352 [==============================] - 398s 294ms/step - loss: 0.0157\n",
            "Epoch 8/10\n",
            "1352/1352 [==============================] - 398s 295ms/step - loss: 0.0153\n",
            "Epoch 9/10\n",
            "1352/1352 [==============================] - 399s 295ms/step - loss: 0.0150\n",
            "Epoch 10/10\n",
            "1352/1352 [==============================] - 398s 294ms/step - loss: 0.0147\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f68061073c8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 58
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uEBDOgoSbVCC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model.save(\"weights.h5\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "K2rqPqPcqh2u",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model.save(\"/content/drive/My Drive/weights.h5\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iJUX31WFqqcS",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "d = model.predict(np.array(gray_l))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tAtpZOycNS2v",
        "colab_type": "code",
        "outputId": "b289d0ef-f590-42bc-d0e6-2747bc47b18a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 285
        }
      },
      "source": [
        "plt.imshow(d[69].reshape(104,104), cmap=\"gray\")"
      ],
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f67ea1d0320>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 72
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQMAAAD7CAYAAABuZ/ELAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRrqxS7sAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cT_-TJRsr9X6",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}