{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "#Original Author: Jonathan Hudson\n", "#CPSC 501 F22" ] }, { "cell_type": "markdown", "metadata": { "id": "DhZTTSlltFh2" }, "source": [ "Imports that are needed" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "id": "KsdVGfVCnQ4J" }, "outputs": [], "source": [ "import sys\n", "import tensorflow as tf\n", "import numpy as np\n", "import matplotlib.pyplot as plt" ] }, { "cell_type": "markdown", "metadata": { "id": "huG1gu0KtG_a" }, "source": [ "Determine the arguments" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "id": "ukCbw8ydnP3k" }, "outputs": [ { "name": "stdin", "output_type": "stream", "text": [ "Dataset: notMNIST\n", "Model: notMNIST.h5\n" ] } ], "source": [ "sys.argv = [\"\", input(\"Dataset:\"), input(\"Model:\")]" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "id": "6TD50Q9e_0ZE" }, "outputs": [], "source": [ "def check_args():\n", " if(len(sys.argv) != 3):\n", " print(\"Usage python predict_test.py \")\n", " sys.exit(1)\n", " if sys.argv[1] == \"MNIST\":\n", " print(\"--Dataset MNIST--\")\n", " class_names = list(range(10))\n", " mnist = tf.keras.datasets.mnist\n", " (x_train, y_train), (x_test, y_test) = mnist.load_data()\n", " x_train, x_test = x_train / 255.0, x_test / 255.0\n", " data = (x_test, y_test)\n", " elif sys.argv[1] == \"notMNIST\":\n", " print(\"--Dataset notMNIST--\")\n", " class_names = [\"A\",\"B\",\"C\",\"D\",\"E\",\"F\",\"G\",\"H\",\"I\",\"J\"]\n", " with np.load(\"notMNIST.npz\", allow_pickle=True) as f:\n", " x_train, y_train = f['x_train'], f['y_train']\n", " x_test, y_test = f['x_test'], f['y_test']\n", " x_train, x_test = x_train / 255.0, x_test / 255.0\n", " data = (x_test, y_test)\n", " else:\n", " print(f\"Choose MNIST or notMNIST, not {sys.argv[1]}\")\n", " sys.exit(2)\n", " if sys.argv[2][-3:] != \".h5\":\n", " print(f\"{sys.argv[2]} is not a h5 extension\")\n", " sys.exit(3)\n", " return class_names, data" ] }, { "cell_type": "markdown", "metadata": { "id": "GFqxpnv4tjR6" }, "source": [ "Completed functions to plot for you" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "id": "E75I4-RVeme5" }, "outputs": [], "source": [ "def plot(class_names, prediction, true_label, predicted_label, img):\n", " plt.figure(figsize=(6,3))\n", " plt.subplot(1,2,1)\n", " plt.grid(False)\n", " plt.xticks([])\n", " plt.yticks([])\n", " plt.imshow(img, cmap=plt.cm.binary)\n", " predicted_label = np.argmax(prediction)\n", " if predicted_label == true_label:\n", " color = 'blue'\n", " else:\n", " color = 'red'\n", " plt.xlabel(\"{} {:2.0f}% ({})\".format(class_names[predicted_label],100*np.max(prediction),class_names[true_label]),color=color)\n", " plt.subplot(1,2,2)\n", " plt.grid(False)\n", " plt.xticks(range(10))\n", " plt.yticks([])\n", " thisplot = plt.bar(class_names, prediction, color=\"#777777\")\n", " plt.ylim([0, 1])\n", " thisplot[predicted_label].set_color('red')\n", " thisplot[true_label].set_color('blue')" ] }, { "cell_type": "markdown", "metadata": { "id": "kL_b5WbYtonX" }, "source": [ "Finish this function to predict output from input" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "GChTgYKGtoFN" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "--Dataset notMNIST--\n", "--Load Model notMNIST.h5--\n" ] }, { "name": "stdin", "output_type": "stream", "text": [ "Pick test_image (0 -> 9999): 1\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "--Should be Class 0--\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdin", "output_type": "stream", "text": [ "Pick test_image (0 -> 9999): 2\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "--Should be Class 5--\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdin", "output_type": "stream", "text": [ "Pick test_image (0 -> 9999): 3\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "--Should be Class 9--\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "def main():\n", " class_names, data = check_args()\n", " x_test, y_test = data\n", " print(f\"--Load Model {sys.argv[2]}--\")\n", " #Load the model that should be in sys.argv[2]\n", " model = None \n", " pick = input(f\"Pick test_image (0 -> {len(x_test)-1}):\")\n", " while pick.isdigit() and int(pick) >= 0 and int(pick) < len(x_test):\n", " pick = int(pick)\n", " img = x_test[pick]\n", " guess = y_test[pick]\n", " print(f\"--Should be Class {guess}--\")\n", " predict(model, class_names, img, guess)\n", " pick = input(f\"Pick test_image (0 -> {len(x_test)-1}):\")\n", " print(\"Done\")\n", "\n", "def predict(model, class_names, img, true_label):\n", " img = np.array([img])\n", " #Replace these two lines with code to make a prediction\n", " prediction = [1/10,1/10,1/10,1/10,1/10,1/10,1/10,1/10,1/10,1/10]\n", " #Determine what the predicted label is\n", " predicted_label = 0\n", " plot(class_names, prediction, true_label, predicted_label, img[0])\n", " plt.show()\n", "main()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "colab": { "collapsed_sections": [], "name": "predict_test.ipynb", "provenance": [] }, "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.7" } }, "nbformat": 4, "nbformat_minor": 4 }