30_langchain_bia_jw.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b97b3b00-8ff4-4e1b-b7c7-709f87aabc37",
   "metadata": {},
   "source": [
    "## Prompting bio-image analysis tasks using LangChain\n",
    "In this notebook we demonstrate how to prompt for executing bio-image analysis tasks using chatGPT and [LangChain](https://github.com/hwchase17/langchain). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f4ae3a80-b6ea-4409-95b7-caecd4e4211c",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from langchain.memory import ConversationBufferMemory\n",
    "from langchain.chat_models import ChatOpenAI\n",
    "from langchain.agents import initialize_agent\n",
    "from langchain.agents import AgentType\n",
    "from langchain.tools import tool\n",
    "\n",
    "from skimage.io import imread\n",
    "# from napari_segment_blobs_and_things_with_membranes import voronoi_otsu_labeling\n",
    "\n",
    "import stackview\n",
    "import numpy as np\n",
    "import napari"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6b78c8e5-58d1-4750-b659-e639a2b99d2f",
   "metadata": {},
   "source": [
    "For accomplishing this, we need an image storage. To keep it simple, we use a dictionary."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8f8158b6-5a36-4cad-a28f-42cd375a0d4f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# global (notebook-wide) variables\n",
    "image_storage = {}\n",
    "viewers = []\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a68ae717-2f4d-4327-b7f1-a4cf13e874d3",
   "metadata": {},
   "source": [
    "To demonstrate bio-image analysis using English language, we define common bio-image analysis functions for loading images, segmenting and counting objects and showing results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a9a8ad4d-7328-465c-8887-cf3fde3d42f1",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "tools = []"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a16cc949-4a0d-4523-af76-50a1806a18ae",
   "metadata": {},
   "source": [
    "@tools.append\n",
    "@tool\n",
    "def tool():\n",
    "    '''\n",
    "\n",
    "    Useful for ...\n",
    "    \n",
    "    '''\n",
    "\n",
    "    return "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "17673761-9801-46c7-a4b5-74743ba8c6af",
   "metadata": {},
   "source": [
    "## napari tools"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a74940b6-c1ad-4170-89c6-fc546852dc01",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tools.append\n",
    "@tool\n",
    "def open_napari_viewer(whatever):\n",
    "    '''\n",
    "\n",
    "    Useful for opening a napari viewer instance.\n",
    "\n",
    "    '''\n",
    "\n",
    "    viewers.append(napari.Viewer())\n",
    "    print(\"I should be starting a napari instance right now... \")\n",
    "\n",
    "    return 'Viewer was opened in a new window.'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "23ee17fa-1b47-4059-a3ba-4db663aab014",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tools.append\n",
    "@tool\n",
    "def display_image_in_napari(image_name):\n",
    "    '''\n",
    "    Useful for showing an image in a napari viewer after the viewer has been opened.\n",
    "\n",
    "    '''\n",
    "    image = image_storage[image_name]\n",
    "    viewers[-1].add_image(image) # use most recent viewer\n",
    "\n",
    "    print(\"I should be adding an image to napari right now...\")\n",
    "\n",
    "    return \"The image was displayed in a napari viewer.\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "71bb0498-88d1-4b7b-91a9-e40a2515eeed",
   "metadata": {},
   "outputs": [],
   "source": [
    "# viewers[0]= napari.Viewer().layers[0].data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "abe838f5-6374-4d38-a008-9052fd436b59",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tools.append\n",
    "@tool\n",
    "def add_two_images(napari_viewer_name):\n",
    "    \"\"\"\n",
    "    Useful for adding two images in a napari viewer\n",
    "    \"\"\"\n",
    "    image1=viewers[-1].layers[-2].data\n",
    "    image2=viewers[-1].layers[-1].data\n",
    "    image_sum = np.add(image1,image2)\n",
    "    viewers[-1].add_image(image_sum)\n",
    "    layer_name = viewers[-1].layers[-1].name\n",
    "    return \"Sum of two images added to a new napari layer\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "bc5b05a7-8ef6-458f-acbf-1c79e26cf9fb",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "@tool\n",
    "def load_image(filename:str):\n",
    "    \"\"\"Useful for loading and image file and storing it.\"\"\"\n",
    "    print(\"loading\", filename)\n",
    "    image = imread(filename)\n",
    "    image_storage[filename] = image\n",
    "    return \"The image is now stored as \" + filename\n",
    "\n",
    "tools.append(load_image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "993a17aa-57b2-4e72-b546-0ec7199c40c6",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# @tool\n",
    "# def segment_bright_objects(image_name):\n",
    "#     \"\"\"Useful for segmenting bright objects in an image that has been loaded and stored before.\"\"\"\n",
    "#     print(\"segmenting\", image_name)\n",
    "    \n",
    "#     image = image_storage[image_name]\n",
    "#     label_image = voronoi_otsu_labeling(image, spot_sigma=4)\n",
    "    \n",
    "#     label_image_name = \"segmented_\" + image_name\n",
    "#     image_storage[label_image_name] = label_image\n",
    "    \n",
    "#     return \"The segmented image has been stored as \" + label_image_name\n",
    "\n",
    "# tools.append(segment_bright_objects)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "a11fe914-4162-4ca3-b067-e5278711e3f3",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "@tool\n",
    "def show_image(image_name):\n",
    "    \"\"\"Useful for showing an image that has been loaded and stored before.\"\"\"\n",
    "    print(\"showing\", image_name)\n",
    "    \n",
    "    image = image_storage[image_name]\n",
    "    display(stackview.insight(image))\n",
    "    \n",
    "    return \"The image \" + image_name + \" is shown above.\"\n",
    "\n",
    "tools.append(show_image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "7e0ce277-682f-4ce2-bb73-bb31b52114d5",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tools.append\n",
    "@tool\n",
    "def move_stage(distance:str):\n",
    "    '''\n",
    "\n",
    "    Useful for moving a microscope stage.\n",
    "    \n",
    "    '''\n",
    "    print(f\"Moving stage....{distance}\")\n",
    "    return f\"The stage was moved {distance}\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "90e514ba-0597-4881-bdcd-49994e4e35a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tools.append\n",
    "@tool\n",
    "def acquire_image(image_name:str):\n",
    "    '''\n",
    "\n",
    "    Useful for snapping an image\n",
    "    \n",
    "    '''\n",
    "    image_storage[image_name]=np.random.random((10,10))\n",
    "    print(f\"Acquiring image ....{image_name}\")\n",
    "    # return f\"{image_name} was acquired.\"\n",
    "    return f\"the image was stored under {image_name}\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "877a4c52-cea0-4a18-b1f4-7a88709713bd",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "@tool\n",
    "def count_objects(image_name):\n",
    "    \"\"\"Useful for counting objects in a segmented image that has been loaded and stored before.\"\"\"\n",
    "    label_image = image_storage[image_name]\n",
    "    \n",
    "    num_labels = label_image.max()\n",
    "    print(\"counting labels in \", image_name, \":\", num_labels)\n",
    "\n",
    "    return f\"The label image {image_name} contains {num_labels} labels.\"\n",
    "\n",
    "tools.append(count_objects)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c0524eb1-7633-45e7-982b-1c2cc5af0b16",
   "metadata": {},
   "source": [
    "We create some memory and a large language model based on OpenAI's chatGPT."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "5d032bf0-49d1-42d4-9654-394a9e660996",
   "metadata": {},
   "outputs": [],
   "source": [
    "memory = ConversationBufferMemory(memory_key=\"chat_history\", return_messages=True)\n",
    "llm=ChatOpenAI(openai_api_key='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', temperature=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7bda4152-8cd8-4257-8e7a-e31fca49ffad",
   "metadata": {
    "tags": []
   },
   "source": [
    "Given the list of tools, the large language model and the memory, we can create an agent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "28afdf8e-87f2-44a7-9f8d-ef188e0f13b5",
   "metadata": {},
   "outputs": [],
   "source": [
    "agent = initialize_agent(\n",
    "    tools, \n",
    "    llm, \n",
    "    agent=AgentType.CHAT_CONVERSATIONAL_REACT_DESCRIPTION, \n",
    "    memory=memory\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23e3065d-8d55-46dc-b160-ff4349ee3beb",
   "metadata": {
    "tags": []
   },
   "source": [
    "This agent can then respond to prompts."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "5bf8d165-de48-4052-8121-d0bedac8a3e2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "loading ./blobs.tif\n",
      "showing ./blobs.tif\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<tr>\n",
       "<td>\n",
       "<img src=\"data:image/png;base64,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\"></img>\n",
       "</td>\n",
       "<td style=\"text-align: center; vertical-align: top;\">\n",
       "\n",
       "<table>\n",
       "<tr><td>shape</td><td>(254,&nbsp;256)</td></tr>\n",
       "<tr><td>dtype</td><td>uint8</td></tr>\n",
       "<tr><td>size</td><td>63.5 kB</td></tr>\n",
       "<tr><td>min</td><td>8</td></tr><tr><td>max</td><td>248</td></tr>\n",
       "</table>\n",
       "<img src=\"data:image/png;base64,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\"></img>\n",
       "</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "StackViewNDArray([[ 40,  32,  24, ..., 216, 200, 200],\n",
       "                  [ 56,  40,  24, ..., 232, 216, 216],\n",
       "                  [ 64,  48,  24, ..., 240, 232, 232],\n",
       "                  ...,\n",
       "                  [ 72,  80,  80, ...,  48,  48,  48],\n",
       "                  [ 80,  80,  80, ...,  48,  48,  48],\n",
       "                  [ 96,  88,  80, ...,  48,  48,  48]], dtype=uint8)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "'The image ./blobs.tif has been loaded and shown.'"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "agent.run(input=\"Please load the image ./blobs.tif and show it.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b67756af-d608-4a24-a202-586425087e60",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Moving stage....10µm\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'The stage was moved 10µm.'"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "agent.run(input=\"Please move the stage 10µm.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "62712464-5dad-4165-888e-52a0ced19687",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Acquiring image ....image1\n",
      "Acquiring image ....image2\n",
      "showing image1\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<tr>\n",
       "<td>\n",
       "<img src=\"data:image/png;base64,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\"></img>\n",
       "</td>\n",
       "<td style=\"text-align: center; vertical-align: top;\">\n",
       "\n",
       "<table>\n",
       "<tr><td>shape</td><td>(10,&nbsp;10)</td></tr>\n",
       "<tr><td>dtype</td><td>float64</td></tr>\n",
       "<tr><td>size</td><td>800.0 B</td></tr>\n",
       "<tr><td>min</td><td>0.009267655911696981</td></tr><tr><td>max</td><td>0.995318200045512</td></tr>\n",
       "</table>\n",
       "<img src=\"data:image/png;base64,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\"></img>\n",
       "</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "StackViewNDArray([[0.03813536, 0.20667607, 0.16002455, 0.6449102 ,\n",
       "                   0.68861675, 0.53737704, 0.21051886, 0.75951651,\n",
       "                   0.88712301, 0.45516098],\n",
       "                  [0.43271903, 0.46176956, 0.94409743, 0.72231517,\n",
       "                   0.42620859, 0.88306326, 0.02046207, 0.0592101 ,\n",
       "                   0.22534072, 0.60006865],\n",
       "                  [0.39464898, 0.33487351, 0.1882166 , 0.13802981,\n",
       "                   0.24796967, 0.91086802, 0.67444452, 0.43195632,\n",
       "                   0.79058648, 0.33769141],\n",
       "                  [0.58440481, 0.35465647, 0.99473701, 0.24195595,\n",
       "                   0.16095799, 0.35838054, 0.82301279, 0.41092173,\n",
       "                   0.56640047, 0.21423328],\n",
       "                  [0.99394216, 0.74493614, 0.206547  , 0.4503094 ,\n",
       "                   0.83291491, 0.29634364, 0.05721871, 0.34956506,\n",
       "                   0.50030899, 0.51629284],\n",
       "                  [0.84718374, 0.2745969 , 0.1216208 , 0.97144285,\n",
       "                   0.73050328, 0.81613631, 0.73722771, 0.05766162,\n",
       "                   0.6876693 , 0.76151074],\n",
       "                  [0.1305059 , 0.42589609, 0.41867167, 0.00926766,\n",
       "                   0.3417662 , 0.057331  , 0.38856756, 0.76313516,\n",
       "                   0.9953182 , 0.8374954 ],\n",
       "                  [0.77899312, 0.04239613, 0.02248155, 0.06798677,\n",
       "                   0.43089875, 0.81819726, 0.08482477, 0.15181204,\n",
       "                   0.08907459, 0.06182703],\n",
       "                  [0.88351541, 0.7113411 , 0.07650285, 0.51428151,\n",
       "                   0.94631159, 0.25937297, 0.6619326 , 0.11090036,\n",
       "                   0.70398772, 0.72257681],\n",
       "                  [0.60879482, 0.69766216, 0.17223723, 0.96126552,\n",
       "                   0.76673338, 0.07853989, 0.07371657, 0.65120294,\n",
       "                   0.03548713, 0.45583821]])"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "showing image2\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<tr>\n",
       "<td>\n",
       "<img src=\"data:image/png;base64,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\"></img>\n",
       "</td>\n",
       "<td style=\"text-align: center; vertical-align: top;\">\n",
       "\n",
       "<table>\n",
       "<tr><td>shape</td><td>(10,&nbsp;10)</td></tr>\n",
       "<tr><td>dtype</td><td>float64</td></tr>\n",
       "<tr><td>size</td><td>800.0 B</td></tr>\n",
       "<tr><td>min</td><td>0.0016860991774779732</td></tr><tr><td>max</td><td>0.9998702283358024</td></tr>\n",
       "</table>\n",
       "<img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAALQAAAB4CAYAAABb59j9AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAACmElEQVR4nO3dQWrbQBiG4ZHpVj6AsaD36YV6hF6o9yk46AAyFLrRdFEcWjeQGmls+evzrOKYzMjwMij8KOlqrbVAiN2jLwDWJGiiCJoogiaKoIkiaKIImiiCJoqgiSJoogiaKIImyodWC8/zXMZxLH3fl67rWm1DgFprOZ/P5XA4lN1u2RnbLOhxHMswDK2WJ9DpdCrH43HRGs2C7vu+lPLrIvf7fattCDBNUxmG4bWZJZoFfbnN2O/3guafrHFr6pdCogiaKM1uOfg/ffz89Y/X3758uuv+TmiiCJoogiaKoIkiaKIImiiCJoqgiSJoogiaKJscff8+Pr336JTn5oQmiqCJImiiCJoogiaKoIkiaKIImiiCJoqgiSJoogiaKIImiqCJImiiCJoogiaKoIkiaKIImiiCJsomn/pO9+xPtW/5+p3QRBE0UQRNFEETRdBEETRRBE0UQRNF0EQRNFE2MfpeOkq9/vmtrbc1732eNT/vvf/3txOaKIImiqCJImiiCJoogiaKoIkiaKIImiiCJsomRt/v2froufUo+b1R/NL97zkKb80JTRRBE0XQRBE0UQRNFEETRdBEETRRBE0UQRPlIaPvtUfBj76etfd/9PXc4vqp7kdzQhNF0EQRNFEETRRBE0XQRBE0UQRNFEETRdBEeYqnvpdq/QfRb32fdpzQRBE0UQRNFEETRdBEETRRBE0UQRNF0ERpNimstZZSSpmm6a/35h/fX7+epsnrlV9fa7nfrd66vsv3Ls0s0dU1VnnDy8tLGYahxdKEOp1O5Xg8LlqjWdDzPJdxHEvf96XruhZbEKLWWs7nczkcDmW3W3YX3CxoeAS/FBJF0EQRNFEETRRBE0XQRBE0UQRNFEETRdBE+Qnl5UP5InA6DgAAAABJRU5ErkJggg==\"></img>\n",
       "</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "StackViewNDArray([[0.32308184, 0.75261688, 0.65544836, 0.44834216,\n",
       "                   0.08185267, 0.61021151, 0.03891287, 0.51795974,\n",
       "                   0.77301545, 0.95434906],\n",
       "                  [0.99987023, 0.95916688, 0.86575117, 0.82425627,\n",
       "                   0.97335635, 0.25541819, 0.56219136, 0.19652452,\n",
       "                   0.0016861 , 0.01759933],\n",
       "                  [0.07335316, 0.19027208, 0.36492408, 0.40423671,\n",
       "                   0.0559174 , 0.12794366, 0.82451812, 0.79166167,\n",
       "                   0.72035382, 0.6093158 ],\n",
       "                  [0.81551482, 0.48801984, 0.08941677, 0.3110388 ,\n",
       "                   0.95990316, 0.89912128, 0.95843673, 0.65065067,\n",
       "                   0.41748763, 0.54537221],\n",
       "                  [0.67560069, 0.04006568, 0.77004796, 0.11833636,\n",
       "                   0.40808656, 0.18679708, 0.18243606, 0.88118289,\n",
       "                   0.96578717, 0.51120727],\n",
       "                  [0.48272731, 0.11446845, 0.86276629, 0.9539643 ,\n",
       "                   0.95701865, 0.98627548, 0.93612211, 0.08371499,\n",
       "                   0.51419511, 0.15883236],\n",
       "                  [0.88950118, 0.36834878, 0.35833936, 0.30498431,\n",
       "                   0.87906558, 0.80366977, 0.41257474, 0.35521912,\n",
       "                   0.04357424, 0.1164521 ],\n",
       "                  [0.52162247, 0.81266623, 0.62886191, 0.72188904,\n",
       "                   0.52168565, 0.68406044, 0.19948448, 0.32327173,\n",
       "                   0.58695366, 0.18350358],\n",
       "                  [0.4944215 , 0.7502539 , 0.8171234 , 0.91532301,\n",
       "                   0.85607533, 0.97566791, 0.3144513 , 0.32945503,\n",
       "                   0.15264128, 0.93938401],\n",
       "                  [0.23872289, 0.09264096, 0.08883559, 0.20325879,\n",
       "                   0.8911192 , 0.57542916, 0.68253247, 0.07786258,\n",
       "                   0.77293817, 0.61614667]])"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "'The images image1 and image2 have been shown.'"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "agent.run(input=\"Take an image, move the stage 10µm, and take another image.  Show both images with the show_image tool.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "56ae45b4-deb5-4d75-ab7b-6e56982c7935",
   "metadata": {},
   "outputs": [],
   "source": [
    "# viewer = napari.Viewer()\n",
    "# viewer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "dcb9d87b-d988-4216-9b3f-88eb54e6c6d7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "I should be starting a napari instance right now... \n",
      "I should be adding an image to napari right now...\n",
      "Moving stage....10µm\n",
      "Acquiring image ....image2\n",
      "I should be adding an image to napari right now...\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'The two images have been added together in the napari viewer.'"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "agent.run(input=\"\"\"\n",
    "    please do the following tasks in this order:\n",
    "    open a napari viewer,\n",
    "    Take an image, \n",
    "    show the image in napari,\n",
    "    move the stage 10µm,\n",
    "    take another image,\n",
    "    show the image in napari,\n",
    "    add the two images in napari.\n",
    "\"\"\")\n",
    "\n",
    "    # use the image math plugin to add the two images in napari."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "16e51b0f-1ee8-4d02-afef-5faff7b20fc0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# agent.run(input=\"\"\"\n",
    "#      use the image math plugin to add the two images in the current napari viewer instance\n",
    "# \"\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ec045ffe-491a-4387-a449-b5318769de03",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}