2-4-CIVITAI-text2img.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/JSJeong-me/AI-Innovation-2024/blob/main/2-4-CIVITAI-text2img.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5a_Cnu2Hw9pJ"
      },
      "source": [
        "# Civitai SDK Text2Img Example\n",
        "This notebook demonstrates how to generate an image using the Civitai SDK."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "O1KGpgzbw9pO"
      },
      "outputs": [],
      "source": [
        "%pip install civitai-py\n",
        "%pip install ipython"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Y6_l9guTw9pQ"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "\n",
        "api_token = input(\"Enter your Civitai API token: \")\n",
        "os.environ['CIVITAI_API_TOKEN'] = api_token"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IS8bGMacw9pR"
      },
      "source": [
        "### Example 1: Manually polling for status of job"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "zPjr-ao_w9pR"
      },
      "outputs": [],
      "source": [
        "# Import the Civitai SDK\n",
        "import civitai\n",
        "\n",
        "# Define the input parameters for the image creation\n",
        "input = {\n",
        "    \"model\": \"urn:air:sdxl:checkpoint:civitai:101055@128078\",\n",
        "    \"params\": {\n",
        "        \"prompt\": \"RAW photo, face portrait photo of woman, wearing black dress, happy face, hard shadows, cinematic shot, dramatic lighting\",\n",
        "        \"negativePrompt\": \"(deformed, distorted, disfigured:1.3)\",\n",
        "        \"scheduler\": \"EulerA\",\n",
        "        \"steps\": 20,\n",
        "        \"cfgScale\": 7,\n",
        "        \"width\": 768,\n",
        "        \"height\": 512,\n",
        "        \"seed\": -1,\n",
        "        \"clipSkip\": 1\n",
        "    },\n",
        "}\n",
        "\n",
        "# Generate the image\n",
        "# We're using await in the notebook to handle the coroutine\n",
        "response = await civitai.image.create(input)\n",
        "print(\"Response:\", response)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Rdrb2T1Sw9pS"
      },
      "source": [
        "You can then retrieve the job status and image using the job token or job ID."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "UchQKy8Qw9pT",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 570
        },
        "outputId": "11c4b41c-af2f-4572-ee0f-2fed47cd07ad"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Job Status Response: {'token': 'eyJKb2JzIjpbImU5MTVhYWRjLTgyNGMtNDI4Ni05MjE0LTNjNzA5YjRlZDQ0MCJdfQ==', 'jobs': [{'jobId': 'e915aadc-824c-4286-9214-3c709b4ed440', 'cost': 0.9600000000000002, 'result': {'blobKey': '82F25F64500A654F9E4503EB460BD219CDCF793DEFB457C0BC16B3455448949A', 'available': True, 'blobUrl': 'https://blobs-temp.sfo3.digitaloceanspaces.com/82F25F64500A654F9E4503EB460BD219CDCF793DEFB457C0BC16B3455448949A?X-Amz-Expires=3600&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=DO00F84RAAYEUTBJ6D9L%2F20240905%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20240905T012617Z&X-Amz-SignedHeaders=host&X-Amz-Signature=c2631dc9cd86f4db1cc12b730e38a185cbaaec1348af44559a8c2e936bcd681b', 'blobUrlExpirationDate': '2024-09-05T02:26:17.3216089Z'}, 'scheduled': False}]}\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "<img src=\"https://blobs-temp.sfo3.digitaloceanspaces.com/82F25F64500A654F9E4503EB460BD219CDCF793DEFB457C0BC16B3455448949A?X-Amz-Expires=3600&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=DO00F84RAAYEUTBJ6D9L%2F20240905%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20240905T012617Z&X-Amz-SignedHeaders=host&X-Amz-Signature=c2631dc9cd86f4db1cc12b730e38a185cbaaec1348af44559a8c2e936bcd681b\"/>"
            ],
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "from IPython.display import Image, display\n",
        "\n",
        "job_token = response['token']\n",
        "job_id = response['jobs'][0]['jobId']\n",
        "\n",
        "# Retrieve job status and image\n",
        "response = await civitai.jobs.get(token=job_token, job_id=job_id)\n",
        "print(\"Job Status Response:\", response)\n",
        "\n",
        "if response['jobs'][0]['result'].get('available'):\n",
        "    image_url = response['jobs'][0]['result'].get('blobUrl')\n",
        "    if image_url:\n",
        "        display(Image(url=image_url))\n",
        "    else:\n",
        "        print(\"Image URL not found in the job result.\")\n",
        "else:\n",
        "    print(\"No image was created, the job is not yet complete, or the result is not available.\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CVS0jLHYw9pT"
      },
      "source": [
        "### Example 2: Run image generation in the background."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RZUxH0mJw9pU"
      },
      "source": [
        "By adding `wait=True` in the `civitai.image.create` call, this runs the text2img job in the background and only returns a response after job completion or timeout. It has a default timeout of 5 minutes.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "j5SZESUTw9pU",
        "outputId": "46f1e429-5b4e-4159-9007-03b1d7a056aa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 570
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Response: {'token': 'eyJKb2JzIjpbImU5MTVhYWRjLTgyNGMtNDI4Ni05MjE0LTNjNzA5YjRlZDQ0MCJdfQ==', 'jobs': [{'jobId': 'e915aadc-824c-4286-9214-3c709b4ed440', 'cost': 0.9600000000000002, 'result': {'blobKey': '82F25F64500A654F9E4503EB460BD219CDCF793DEFB457C0BC16B3455448949A', 'available': True, 'blobUrl': 'https://blobs-temp.sfo3.digitaloceanspaces.com/82F25F64500A654F9E4503EB460BD219CDCF793DEFB457C0BC16B3455448949A?X-Amz-Expires=3600&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=DO00F84RAAYEUTBJ6D9L%2F20240905%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20240905T012610Z&X-Amz-SignedHeaders=host&X-Amz-Signature=4cb80c99125f247dd9e207f3e87f8c38178d3f352aafc5cb2a0a785e75e96d6d', 'blobUrlExpirationDate': '2024-09-05T02:26:10.9024135Z'}, 'scheduled': False}]}\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "<img src=\"https://blobs-temp.sfo3.digitaloceanspaces.com/82F25F64500A654F9E4503EB460BD219CDCF793DEFB457C0BC16B3455448949A?X-Amz-Expires=3600&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=DO00F84RAAYEUTBJ6D9L%2F20240905%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20240905T012610Z&X-Amz-SignedHeaders=host&X-Amz-Signature=4cb80c99125f247dd9e207f3e87f8c38178d3f352aafc5cb2a0a785e75e96d6d\"/>"
            ],
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Generate the image and wait for completion\n",
        "response = await civitai.image.create(input, wait=True)\n",
        "print(\"Response:\", response)\n",
        "\n",
        "if response['jobs'][0]['result'].get('available'):\n",
        "    image_url = response['jobs'][0]['result'].get('blobUrl')\n",
        "    if image_url:\n",
        "        display(Image(url=image_url))\n",
        "    else:\n",
        "        print(\"Image URL not found in the job result.\")\n",
        "else:\n",
        "    print(\"No image was created, the job is not yet complete, or the result is not available.\")"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## LORA 활용"
      ],
      "metadata": {
        "id": "u_8_klGrySkJ"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "1AJLmwDQw9pU"
      },
      "outputs": [],
      "source": [
        "input = {\n",
        "    \"model\": \"urn:air:sd1:checkpoint:civitai:4384@128713\",\n",
        "    \"params\": {\n",
        "        \"prompt\": \"masterpiece, best quality, 1girl, IncrsAhri, multiple tails, fox tail, korean clothes, skirt, braid, arms behind back\",\n",
        "        \"negativePrompt\": \"(worst quality:1.4), (low quality:1.4), simple background, bad anatomy\",\n",
        "        \"scheduler\": \"EulerA\",\n",
        "        \"steps\": 25,\n",
        "        \"cfgScale\": 7,\n",
        "        \"width\": 512,\n",
        "        \"height\": 768,\n",
        "        \"seed\": -1,\n",
        "        \"clipSkip\": 2\n",
        "    },\n",
        "    \"additionalNetworks\": {\n",
        "        \"urn:air:sd1:lora:civitai:162141@182559\": {\n",
        "            \"type\": \"Lora\",\n",
        "            \"strength\": 1.0\n",
        "        }\n",
        "    }\n",
        "}"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Generate the image and wait for completion\n",
        "response = await civitai.image.create(input, wait=True)\n",
        "print(\"Response:\", response)\n",
        "\n",
        "if response['jobs'][0]['result'].get('available'):\n",
        "    image_url = response['jobs'][0]['result'].get('blobUrl')\n",
        "    if image_url:\n",
        "        display(Image(url=image_url))\n",
        "    else:\n",
        "        print(\"Image URL not found in the job result.\")\n",
        "else:\n",
        "    print(\"No image was created, the job is not yet complete, or the result is not available.\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mdIzDHwGylGb",
        "outputId": "0b37a6aa-0017-4793-d36b-66e8360f2c43"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:root:Job eyJKb2JzIjpbIjMwMDQ0ZmJiLTY5MjAtNGRkZS1hMzVmLTZkMzNhODJmNGQxMiJdfQ== did not complete within 5 minutes.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Response: {'token': 'eyJKb2JzIjpbIjMwMDQ0ZmJiLTY5MjAtNGRkZS1hMzVmLTZkMzNhODJmNGQxMiJdfQ==', 'jobs': [{'jobId': '30044fbb-6920-4dde-a35f-6d33a82f4d12', 'cost': 1.2000000000000002, 'result': {'blobKey': '234E5D9DA6930D3E5D41BF082D51DD2E7906B53021E519EBA8953BBE853A2ECF', 'available': False}, 'scheduled': True}]}\n",
            "No image was created, the job is not yet complete, or the result is not available.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "jtj8aFSTyrve"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.12.2"
    },
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "include_colab_link": true
    },
    "accelerator": "GPU"
  },
  "nbformat": 4,
  "nbformat_minor": 0
}