'training_convnet'

SmBito · Oct 15, 2019 · 508b934 · 508b934
1 parent 2fd493e
commit 508b934
Show file tree

Hide file tree

Showing 2 changed files with 178 additions and 15 deletions.
diff --git a/.ipynb_checkpoints/05_Convolutional_Neural_Networks-checkpoint.ipynb b/.ipynb_checkpoints/05_Convolutional_Neural_Networks-checkpoint.ipynb
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -16,16 +16,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [],
    "source": [
     "REBUILD_DATA = True # set to true to one once, then back to false unless you want to change something in your training data.\n",
     "\n",
     "class DogsVSCats():\n",
     "    IMG_SIZE = 50\n",
-    "    CATS = \"/Users/Mac/Downloads/kagglecatsanddogs/PetImages/Cat\"\n",
-    "    DOGS = \"/Users/Mac/Downloads/kagglecatsanddogs/PetImages/Dog\"\n",
+    "    CATS = \"/Users/Mac/Downloads/datasets/kagglecatsanddogs/PetImages/Cat\"\n",
+    "    DOGS = \"/Users/Mac/Downloads/datasets/kagglecatsanddogs/PetImages/Dog\"\n",
     "    TESTING = \"PetImages/Testing\"\n",
     "    LABELS = {CATS: 0, DOGS: 1}\n",
     "    training_data = []\n",
@@ -62,43 +62,43 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "  1%|          | 73/12501 [00:00<00:17, 727.44it/s]"
+      "  0%|          | 0/12501 [00:00<?, ?it/s]"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "/Users/Mac/Downloads/kagglecatsanddogs/PetImages/Cat\n"
+      "/Users/Mac/Downloads/datasets/kagglecatsanddogs/PetImages/Cat\n"
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "100%|██████████| 12501/12501 [02:44<00:00, 75.80it/s]\n",
-      "  0%|          | 8/12501 [00:00<02:49, 73.78it/s]"
+      "100%|██████████| 12501/12501 [03:06<00:00, 66.97it/s]\n",
+      "  0%|          | 5/12501 [00:00<04:16, 48.64it/s]"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "/Users/Mac/Downloads/kagglecatsanddogs/PetImages/Dog\n"
+      "/Users/Mac/Downloads/datasets/kagglecatsanddogs/PetImages/Dog\n"
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "100%|██████████| 12501/12501 [02:22<00:00, 87.59it/s] \n"
+      "100%|██████████| 12501/12501 [02:26<00:00, 85.25it/s] \n"
      ]
     },
     {
@@ -116,12 +116,153 @@
     "    dogsvcats.make_training_data()"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "training_data = np.load(\"training_data.npy\", allow_pickle = True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "24946\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(len(training_data))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[array([[ 60,  64,  69, ..., 198,  47,  43],\n",
+      "       [ 63,  68,  73, ..., 205,  49,  45],\n",
+      "       [ 66,  71,  82, ..., 207, 101,  69],\n",
+      "       ...,\n",
+      "       [ 79, 103, 230, ...,  57,  59,  58],\n",
+      "       [ 75,  33, 225, ...,  72,  71,  69],\n",
+      "       [ 67,  67, 217, ...,  64,  72,  68]], dtype=uint8)\n",
+      " array([0., 1.])]\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(training_data[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[array([[135, 132, 145, ...,  39,  65,  64],\n",
+      "       [100, 135, 130, ...,  43,  47,  45],\n",
+      "       [120, 110, 122, ...,  39,  48,  52],\n",
+      "       ...,\n",
+      "       [ 49,  59,  41, ...,  57, 178, 113],\n",
+      "       [ 77, 141,  69, ..., 132, 131, 136],\n",
+      "       [ 59,  51,  99, ..., 168,  89,  73]], dtype=uint8)\n",
+      " array([0., 1.])]\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(training_data[1])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "plt.imshow(training_data[2][0], cmap = 'gray')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([0., 1.])"
+      ]
+     },
+     "execution_count": 31,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "training_data[2][1]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Trainig Convent"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "import torch\n",
+    "import torch.nn as nn\n",
+    "import torch.nn.functional as F\n",
+    "\n",
+    "class Net(nn.Module):\n",
+    "    def __init__(self):\n",
+    "        super().__init__()\n",
+    "        self.conv1 = nn.Conv2d(1, 32, 5) \n",
+    "        ## input: 1\n",
+    "        ## output: 32\n",
+    "        ## kernel size : 5\n",
+    "        self.conv2 = nn.Conv2d(1, 32, 5) \n",
+    "        self.conv3 = nn.Conv2d(1, 32, 5) \n",
+    "        "
+   ]
   }
  ],
  "metadata": {
@@ -140,7 +281,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.8"
+   "version": "3.7.3"
   }
  },
  "nbformat": 4,