{"cells": [{"cell_type": "markdown", "id": "302934307671667531413257853548643485645", "metadata": {}, "source": ["# Gradio Demo: digit_classifier"]}, {"cell_type": "code", "execution_count": null, "id": "272996653310673477252411125948039410165", "metadata": {}, "outputs": [], "source": ["!pip install -q gradio tensorflow"]}, {"cell_type": "code", "execution_count": null, "id": "288918539441861185822528903084949547379", "metadata": {}, "outputs": [], "source": ["from urllib.request import urlretrieve\n", "\n", "import tensorflow as tf\n", "\n", "import gradio as gr\n", "\n", "urlretrieve(\n", "    \"https://gr-models.s3-us-west-2.amazonaws.com/mnist-model.h5\", \"mnist-model.h5\"\n", ")\n", "model = tf.keras.models.load_model(\"mnist-model.h5\")\n", "\n", "\n", "def recognize_digit(image):\n", "    image = image.reshape(1, -1)\n", "    prediction = model.predict(image).tolist()[0]\n", "    return {str(i): prediction[i] for i in range(10)}\n", "\n", "\n", "im = gr.Image(shape=(28, 28), image_mode=\"L\", invert_colors=False, source=\"canvas\")\n", "\n", "demo = gr.Interface(\n", "    recognize_digit,\n", "    im,\n", "    gr.Label(num_top_classes=3),\n", "    live=True,\n", "    capture_session=True,\n", ")\n", "\n", "if __name__ == \"__main__\":\n", "    demo.launch()\n"]}], "metadata": {}, "nbformat": 4, "nbformat_minor": 5}