{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "gradio_demos.ipynb",
      "private_outputs": true,
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyNhpeeMxUCUtvCsAPt4I1BZ",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/nlp-en-es/nlp-de-cero-a-cien/blob/main/6_demos/gradio_demos.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z5L1Z2AcwUPX"
      },
      "source": [
        "!pip install transformers gradio sentencepiece"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GC1HwQjqQRjS"
      },
      "source": [
        "# Example 1. Normal inference of a model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AHRKQwWuwrw3"
      },
      "source": [
        "import gradio as gr\n",
        "\n",
        "from transformers import pipeline\n",
        "\n",
        "pipe = pipeline(\"translation\", model=\"Helsinki-NLP/opus-mt-en-es\")\n",
        "\n",
        "def predict(text):\n",
        "  return pipe(text)[0][\"translation_text\"]\n",
        "  \n",
        "title = \"Interactive demo: Helsinki-NLP English to Spanish Translation\"\n",
        "\n",
        "iface = gr.Interface(\n",
        "  fn=predict, \n",
        "  inputs=[gr.inputs.Textbox(label=\"text\", lines=3)],\n",
        "  outputs='text',\n",
        "  title=title,\n",
        "  examples=[[\"Hello! My name is Omar\"], [\"I like this workshop\"]]\n",
        ")\n",
        "\n",
        "iface.launch(debug=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PwTqI-t0QZtM"
      },
      "source": [
        "# Example 2. Using Inference API from Hugging Face"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EDkCXgXBx_MU"
      },
      "source": [
        "import gradio as gr\n",
        "\n",
        "description = \"BigGAN text-to-image demo.\"\n",
        "title = \"BigGAN ImageNet\"\n",
        "\n",
        "interface = gr.Interface.load(\n",
        "    \"huggingface/osanseviero/BigGAN-deep-128\", \n",
        "    description=description,\n",
        "    title = title,\n",
        "    examples=[[\"american robin\"], [\"chest\"], [\"soap bubble\"]]\n",
        ")\n",
        "\n",
        "interface.launch()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WffHZhIzURoW"
      },
      "source": [
        "# Example 3. Using series of models"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sS6iAMYKQpkH"
      },
      "source": [
        "import gradio as gr\n",
        "from gradio.mix import Series\n",
        "\n",
        "description = \"Generate your own D&D story!\"\n",
        "title = \"Spanish Story Generator using Opus MT and GPT-2\"\n",
        "\n",
        "translator_es = gr.Interface.load(\"huggingface/Helsinki-NLP/opus-mt-es-en\")\n",
        "story_gen = gr.Interface.load(\"huggingface/pranavpsv/gpt2-genre-story-generator\")\n",
        "translator_en = gr.Interface.load(\"huggingface/Helsinki-NLP/opus-mt-en-es\")\n",
        "\n",
        "examples = [[\"La aventura comienza en\"]]\n",
        "\n",
        "interface = Series(translator_es, story_gen, translator_en, description = description,\n",
        "      title = title,\n",
        "      examples=examples, \n",
        "      inputs = gr.inputs.Textbox(lines = 10)\n",
        ")\n",
        "\n",
        "interface.launch()"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}