intro_demo.ipynb

{
 "cells": [
  {
   "attachments": {
    "ff162b92-925b-4d7a-a719-73103e5981d8.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "9c4ba68a-ae5e-45cc-8202-17011b7572bd",
   "metadata": {},
   "source": [
    "<div>\n",
    "<img src=\"attachment:ff162b92-925b-4d7a-a719-73103e5981d8.png\" width=\"600\"/>\n",
    "</div>\n",
    "\n",
    "# Introduction to `fastplotlib`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7b1a583-a830-42d4-a4fa-17bac94537db",
   "metadata": {},
   "source": [
    "## `fastplotlib` API\n",
    "\n",
    "### 1. Graphics - objects that are drawn\n",
    "- `Image`, `Line`, `Scatter`\n",
    "- Collections - `LineCollection`, `LineStack` (ex: neural timeseries data)\n",
    "\n",
    "    #### a) Graphic Properties\n",
    "  - Common: `Name`, `Offset`, `Rotation`, `Visible`, `Deleted`\n",
    "  - Graphic Specific: `ImageVmin`, `ImageVmax`, `VertexColors`, etc.\n",
    "\n",
    "  #### b) Selectors\n",
    "  - `LinearSelector` - horizontal or vertical line slider\n",
    "  - `LinearRegionSelector` - horizontal or vertical resizable region selection\n",
    "\n",
    "### 2. Layouts\n",
    "- `Figure` - a single plot or a grid of subplots\n",
    "\n",
    "### 3. Widgets - high level widgets to make repetitive UIs easier\n",
    "- `ImageWidget`- widget for `ImageGraphic` data with dims: `xy`, `txy`, `tzxy`\n",
    "- Sliders, support window functions, etc."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e443dcc4-8d1d-4eed-8b04-aa358181d3e5",
   "metadata": {},
   "source": [
    "This notebook will go through some basic components of the `fastplotlib` API including how to instantiate a plot, add graphics to a plot, and subsequently interact with the plot. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7133caab-f840-4719-8d3f-b1a81718d7fa",
   "metadata": {},
   "source": [
    "**The example images are from `imageio` so you will need to install it for this example notebook. But `imageio` is not required to use fasptlotlib**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "52c12cb2-1b3c-4578-81d6-b64b7598ffec",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install imageio"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "bdc431ce-9c99-4c8e-b925-f8136c4778cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "import imageio.v3 as iio"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1943949e-1aee-48ec-a280-d743e41e7f15",
   "metadata": {},
   "source": [
    "By default, `fastplotlib` will enumerate the available devices and highlight which device has been selected by default when importing the module."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c3ceed4f-a32c-489a-9cad-f7cf7d188662",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e36aa378c98141ab949e9d6dcf16791b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Image(value=b'\\x89PNG\\r\\n\\x1a\\n\\x00\\x00\\x00\\rIHDR\\x00\\x00\\x01,\\x00\\x00\\x007\\x08\\x06\\x00\\x00\\x00\\xb6\\x1bw\\x99\\x…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Unable to find extension: VK_EXT_swapchain_colorspace\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Available devices:\n",
      "🯄 | AMD Radeon Graphics (RADV REMBRANDT) | IntegratedGPU | Vulkan | Mesa 22.3.6\n",
      "✅ (default) | NVIDIA T1200 Laptop GPU | DiscreteGPU | Vulkan | 535.129.03\n",
      "❗ | llvmpipe (LLVM 15.0.6, 256 bits) | CPU | Vulkan | Mesa 22.3.6 (LLVM 15.0.6)\n",
      "❗ | AMD Radeon Graphics (rembrandt, LLVM 15.0.6, DRM 3.52, 6.4.0-0.deb12.2-amd64) | Unknown | OpenGL | \n"
     ]
    }
   ],
   "source": [
    "import fastplotlib as fpl\n",
    "import numpy as np\n",
    "from itertools import product"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a65f161-9018-4439-a124-0c301c4850f2",
   "metadata": {},
   "source": [
    "## Simple image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "602a6498-d4ed-4b5f-80b8-1f6db5b59460",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c258251b26924cb6841e9efe2c18c3cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "RFBOutputContext()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "JupyterOutputContext(children=(JupyterWgpuCanvas(css_height='500px', css_width='600px'), IpywidgetToolBar(children=(Button(icon='expand-arrows-alt', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='auto-scale scene'), Button(icon='align-center', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='auto-center scene'), ToggleButton(value=True, icon='hand-pointer', layout=Layout(width='auto'), tooltip='panzoom controller'), ToggleButton(value=True, description='1:1', layout=Layout(width='auto'), style=ToggleButtonStyle(font_weight='bold'), tooltip='maintain aspect'), Button(icon='arrow-down', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='y-axis direction'), Button(icon='draw-polygon', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='add PolygonSelector'), ToggleButton(value=False, icon='video', layout=Layout(width='auto'), tooltip='record'), Dropdown(description='Subplots:', layout=Layout(width='200px'), options=('(0, 0)',), value='(0, 0)')))))"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create a `Figure` instance\n",
    "# by default the figure will have 1 subplot\n",
    "fig = fpl.Figure(size=(600, 500))\n",
    "\n",
    "# get a grayscale image\n",
    "data = iio.imread(\"imageio:camera.png\")\n",
    "\n",
    "# plot the image data\n",
    "image_graphic = fig[0, 0].add_image(data=data, name=\"sample-image\")\n",
    "\n",
    "# show the plot\n",
    "fig.show(sidecar=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea639d90-de06-47bf-a7df-7f829efc69b9",
   "metadata": {},
   "source": [
    "**Use the handle on the bottom right corner of the canvas to resize it. You can also pan and zoom using your mouse!**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f4147273-c042-4251-8e1b-2cb9f179a845",
   "metadata": {},
   "source": [
    "Changing graphic **properties**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "94ab4863-aa46-472d-a32a-19617dc24917",
   "metadata": {},
   "outputs": [],
   "source": [
    "image_graphic.cmap = \"viridis\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95ff4e57-dbe2-4f56-8d0a-8e1f56027197",
   "metadata": {},
   "source": [
    "### Slicing data\n",
    "\n",
    "**Most properties, such as data, support slicing!**\n",
    "\n",
    "Our image data is of shape [n_rows, n_columns]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "cca79223-30c8-4e67-800c-d2d96f6a59ca",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(512, 512)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "image_graphic.data.value.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "4bac341a-d89b-41ba-92be-7b87abc2e6f5",
   "metadata": {},
   "outputs": [],
   "source": [
    "image_graphic.data[::15, :] = 1\n",
    "image_graphic.data[:, ::15] = 1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe016bd9-9cb2-481a-bcce-1c85b546c6da",
   "metadata": {},
   "source": [
    "**Fancy indexing**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "289a9df8-c218-4de8-922b-51a314674bf5",
   "metadata": {},
   "outputs": [],
   "source": [
    "image_graphic.data[data > 175] = 255"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "34f15b06-5b1b-4e05-ae0f-4089eed15706",
   "metadata": {},
   "source": [
    "**Adjust vmin vmax**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c567c61e-9475-4036-aaec-b668c74ca676",
   "metadata": {},
   "outputs": [],
   "source": [
    "image_graphic.vmin = 50\n",
    "image_graphic.vmax = 150"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "1295d135-482d-447c-aaf3-96820455e201",
   "metadata": {},
   "outputs": [],
   "source": [
    "image_graphic.reset_vmin_vmax()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f4b8098b-0477-467b-8042-6373033a3d61",
   "metadata": {},
   "source": [
    "### Change graphic properties \n",
    "\n",
    "Now that we have a better idea of how graphic properties can be dynamically changed, let's practice :D\n",
    "\n",
    "**Question:** Can you change the data of the image to be random data of the same shape? Hint: Use `np.random.rand()` to create the new data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "eb989954-c5b8-4d56-8796-b7bd1d47001d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# create new random data the same shape as the original image\n",
    "### insert code\n",
    "new_data = np.random.rand(*image_graphic.data.value.shape)\n",
    "\n",
    "image_graphic.data = new_data\n",
    "# update the data \n",
    "### insert code"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5745de39-70a8-4503-aca9-417d70db31a5",
   "metadata": {},
   "source": [
    "**You should see a solid image. You will need to reset the vmin/vmax of the image.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "18f7cb43-3e2d-493a-a0aa-4cd4a7efc948",
   "metadata": {},
   "outputs": [],
   "source": [
    "# reset the vmin/vmax\n",
    "### insert code here\n",
    "image_graphic.reset_vmin_vmax()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9655cbd-7157-474b-9f73-a5db43930d20",
   "metadata": {},
   "source": [
    "**Question:** Can you change the colormap of the image? See [here](https://matplotlib.org/stable/gallery/color/colormap_reference.html) for a list of colormaps."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "b6f705c6-6616-4f93-a4a6-5b24bb731958",
   "metadata": {},
   "outputs": [],
   "source": [
    "### insert code here\n",
    "image_graphic.cmap = \"hsv\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "8f4a8b9d-9817-43ef-815f-276ff36f6171",
   "metadata": {},
   "outputs": [],
   "source": [
    "# close the plot\n",
    "fig.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81e4114c-3241-44aa-9e8c-156c171cf320",
   "metadata": {},
   "source": [
    "### Image updates\n",
    "\n",
    "This examples show how you can define animation functions that run on every render cycle."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "39e4b3a4-3975-478e-8470-74b4cbe3522f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "46aa562e93cf49ae9ad32179774800f2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "RFBOutputContext()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "JupyterOutputContext(children=(JupyterWgpuCanvas(css_height='500px', css_width='600px'), IpywidgetToolBar(children=(Button(icon='expand-arrows-alt', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='auto-scale scene'), Button(icon='align-center', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='auto-center scene'), ToggleButton(value=True, icon='hand-pointer', layout=Layout(width='auto'), tooltip='panzoom controller'), ToggleButton(value=True, description='1:1', layout=Layout(width='auto'), style=ToggleButtonStyle(font_weight='bold'), tooltip='maintain aspect'), Button(icon='arrow-down', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='y-axis direction'), Button(icon='draw-polygon', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='add PolygonSelector'), ToggleButton(value=False, icon='video', layout=Layout(width='auto'), tooltip='record'), Dropdown(description='Subplots:', layout=Layout(width='200px'), options=('(0, 0)',), value='(0, 0)')))))"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create another `Figure` instance\n",
    "fig_v = fpl.Figure(size=(600, 500))\n",
    "\n",
    "fig.canvas.max_buffered_frames = 1\n",
    "\n",
    "# make some random data again\n",
    "data = np.random.rand(512, 512)\n",
    "\n",
    "# plot the data\n",
    "fig_v[0,0].add_image(data=data, name=\"random-image\")\n",
    "\n",
    "# a function to update the image_graphic\n",
    "# a plot will pass its plot instance to the animation function as an argument\n",
    "\n",
    "def update_data(plot_instance):\n",
    "    new_data = np.random.rand(512, 512)\n",
    "    plot_instance[0,0][\"random-image\"].data = new_data\n",
    "\n",
    "#add this as an animation function\n",
    "fig_v.add_animations(update_data)\n",
    "\n",
    "fig_v.show(sidecar=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "ad014d22-5618-48d8-a4b0-894284df5556",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig_v.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c506b64-7f7b-4ab9-9430-2c04a259b64a",
   "metadata": {},
   "source": [
    "### Image Practice \n",
    "\n",
    "**Question:** Can you do the following:\n",
    "\n",
    "- create a new plot called `practice_fig`\n",
    "\n",
    "- add an `ImageGraphic` with the following characteristics to the figure:\n",
    "\n",
    "  - random data in the shape (512, 512)\n",
    "  - colormap \"viridis\"\n",
    "  - name \"random-image\"\n",
    "\n",
    "- set the top-left and bottom-right quadrants of the data to 1 using slicing :D\n",
    "\n",
    "- add an animation function that updates the top-right and bottom-left quadrants with new random data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "3ab5b823-bd5b-42f7-a1e0-ab4a78cb7aff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2de5f107f37f40cd887654936c14e202",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "RFBOutputContext()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# create a new fig\n",
    "### insert code here\n",
    "practice_fig = fpl.Figure()\n",
    "\n",
    "# create random data in the shape (512, 512)\n",
    "### insert code here\n",
    "data = np.random.rand(512,512)\n",
    "\n",
    "# add image to plot\n",
    "### insert code here\n",
    "image_graphic = practice_fig[0,0].add_image(data=data, cmap=\"viridis\", name=\"random-image\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "6098b032-362b-494c-938e-27e159d822ce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "JupyterOutputContext(children=(JupyterWgpuCanvas(), IpywidgetToolBar(children=(Button(icon='expand-arrows-alt'…"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# show the plot\n",
    "### insert code here\n",
    "practice_fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "cef0d2fe-20a0-4f97-abaa-fa976ecedb5e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# set the top-left and bottom-right quadrants of the data to 1\n",
    "### insert code here\n",
    "practice_fig[0,0][\"random-image\"].data[:256, :256] = 1\n",
    "practice_fig[0,0][\"random-image\"].data[256:, 256:] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "941ab3e5-cbfb-48dd-a237-b83b57008c2d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# define an animation function to update the top-right and bottom-left quadrants with new random data\n",
    "### insert code here\n",
    "def update_data(plot_instance):\n",
    "    new_data = np.random.rand(256, 256)\n",
    "    plot_instance[0,0][\"random-image\"].data[:256, 256:] = new_data\n",
    "    plot_instance[0,0][\"random-image\"].data[256:, :256] = new_data\n",
    "# add the animation function to the figure\n",
    "### insert code here\n",
    "\n",
    "practice_fig.add_animations(update_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "bfd121ce-584b-447a-a155-d9188e64b316",
   "metadata": {},
   "outputs": [],
   "source": [
    "# close the plot\n",
    "### insert code here\n",
    "practice_fig.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dee89adf-f9a7-41b1-bc68-6eb50c04fb8b",
   "metadata": {},
   "source": [
    "## 2D line plots\n",
    "\n",
    "This example plots a sine wave, cosine wave, and ricker wavelet and demonstrates how **Graphic Features** can be modified by slicing!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "edee776a-ccaa-4494-8564-97c37b4a04da",
   "metadata": {},
   "source": [
    "### First generate some data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "31516018-632d-41fc-9e13-cb88a6159b8d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# linspace, create 100 evenly spaced x values from -10 to 10\n",
    "xs = np.linspace(-10, 10, 100)\n",
    "# sine wave\n",
    "ys = np.sin(xs)\n",
    "sine = np.column_stack([xs, ys])\n",
    "\n",
    "# cosine wave\n",
    "ys = np.cos(xs)\n",
    "cosine = np.column_stack([xs, ys])\n",
    "\n",
    "# sinc function\n",
    "a = 0.5\n",
    "ys = np.sinc(xs) * 3\n",
    "sinc = np.column_stack([xs, ys])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32660279-4d41-4247-a1f0-f7cb2eb44e24",
   "metadata": {},
   "source": [
    "### We will plot all of it on the same plot. Each line plot will be an individual `Graphic`, you can have any combination of graphics on a plot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "72e36685-ebf8-4691-b2c1-271c9bf6eda3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c97bd85cd21745a2b9032868b890b7d1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "RFBOutputContext()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/kushal/venvs/workshop/lib/python3.11/site-packages/fastplotlib/graphics/_features/_base.py:21: UserWarning: casting float64 array to float32\n",
      "  warn(f\"casting {array.dtype} array to float32\")\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "JupyterOutputContext(children=(JupyterWgpuCanvas(), IpywidgetToolBar(children=(Button(icon='expand-arrows-alt', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='auto-scale scene'), Button(icon='align-center', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='auto-center scene'), ToggleButton(value=True, icon='hand-pointer', layout=Layout(width='auto'), tooltip='panzoom controller'), ToggleButton(value=True, description='1:1', layout=Layout(width='auto'), style=ToggleButtonStyle(font_weight='bold'), tooltip='maintain aspect'), Button(icon='arrow-up', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='y-axis direction'), Button(icon='draw-polygon', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='add PolygonSelector'), ToggleButton(value=False, icon='video', layout=Layout(width='auto'), tooltip='record'), Dropdown(description='Subplots:', layout=Layout(width='200px'), options=('(0, 0)',), value='(0, 0)')))))"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a figure\n",
    "fig_lines = fpl.Figure()\n",
    "# we will add all the lines to the same subplot\n",
    "subplot = fig_lines[0, 0]\n",
    "\n",
    "# plot sine wave, use a single color\n",
    "sine_graphic = subplot.add_line(data=sine, thickness=5, colors=\"magenta\")\n",
    "\n",
    "# you can also use colormaps for lines!\n",
    "cosine_graphic = subplot.add_line(data=cosine, thickness=12, cmap=\"autumn\", offset=(0, 5, 0))\n",
    "\n",
    "# or a list of colors for each datapoint\n",
    "colors = [\"r\"] * 25 + [\"purple\"] * 25 + [\"y\"] * 25 + [\"b\"] * 25\n",
    "sinc_graphic = subplot.add_line(data=sinc, thickness=5, colors=colors, offset=(0, 8, 0))\n",
    "\n",
    "# show the plot\n",
    "fig_lines.show(sidecar=True, sidecar_kwargs={\"title\": \"lines\"})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "22981996-2c88-4b5c-aaa5-e6e1caa0974f",
   "metadata": {},
   "source": [
    "### Graphic properties support slicing! :D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "d63d971e-2d08-4e9d-81ed-eb8df6de0a8f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# indexing of colors\n",
    "cosine_graphic.colors[:15] = \"magenta\"\n",
    "cosine_graphic.colors[90:] = \"red\"\n",
    "cosine_graphic.colors[60] = \"w\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "ef97df2e-1862-4f03-8286-e049e2e06038",
   "metadata": {},
   "outputs": [],
   "source": [
    "# indexing to assign colormaps to entire lines or segments\n",
    "sinc_graphic.cmap[10:50] = \"gray\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "31030051-b7d3-4d3a-991b-d68b099c5067",
   "metadata": {},
   "outputs": [],
   "source": [
    "# use a qualitative cmap (https://matplotlib.org/stable/_images/sphx_glr_colormaps_006.png)\n",
    "sine_graphic.cmap = \"plasma\"\n",
    "# set the cmap transform \n",
    "sine_graphic.cmap.transform = [4] * 25 + [1] * 25 + [3] * 25 + [8] * 25"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "99e25363-03dd-449a-b648-7d5ee628105d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# close the plot\n",
    "fig_lines.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b72c3c5-b430-4876-8e1e-173a84b54e7d",
   "metadata": {},
   "source": [
    "## `ImageWidget`\n",
    "\n",
    "Great for looking at multi-dimensional image data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "53896982-33c9-480e-a2c0-fdcdf485f816",
   "metadata": {},
   "outputs": [],
   "source": [
    "# load subsampled snippet of zebrafish data from Martin Haesemeyer\n",
    "zfish_data = np.load(\"zfish_test.npy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "75d72707-d641-4960-9441-286105be36eb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "26214400"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "zfish_data.nbytes "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "7277575b-195d-484a-bf3b-923ef4a9a990",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100, 4, 256, 256)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# data is tzxy\n",
    "zfish_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "d25bb23f-9d03-40f8-82c7-9770388dde57",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get the number of planes\n",
    "n_planes = zfish_data.shape[1]\n",
    "n_planes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "680a7202-550c-4b26-a35c-72a4e4caea29",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b8d1ad9cb08f49408f699836ec3d8355",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "RFBOutputContext()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "iw_zfish = fpl.ImageWidget(\n",
    "    data=[zfish_data[:, i] for i in range(n_planes)],\n",
    "    window_funcs={\"t\": (np.mean, 5)},\n",
    "    names=[f\"plane-{i}\" for i in range(n_planes)],\n",
    "    cmap=\"gnuplot2\", \n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "60f3a1af-9293-4e81-9bc0-30010cd8dd4a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "JupyterOutputContext(children=(JupyterWgpuCanvas(), IpywidgetToolBar(children=(Button(icon='expand-arrows-alt', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='auto-scale scene'), Button(icon='align-center', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='auto-center scene'), ToggleButton(value=True, icon='hand-pointer', layout=Layout(width='auto'), tooltip='panzoom controller'), ToggleButton(value=True, description='1:1', layout=Layout(width='auto'), style=ToggleButtonStyle(font_weight='bold'), tooltip='maintain aspect'), Button(icon='arrow-down', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='y-axis direction'), Button(icon='draw-polygon', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='add PolygonSelector'), ToggleButton(value=False, icon='video', layout=Layout(width='auto'), tooltip='record'), Dropdown(description='Subplots:', layout=Layout(width='200px'), options=('plane-0', 'plane-1', 'plane-2', 'plane-3'), value='plane-0'))), IpywidgetImageWidgetToolbar(children=(HBox(children=(Button(icon='adjust', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='reset vmin/vmax'), Button(description='reset', icon='adjust', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='reset vmin/vmax and reset histogram using current frame'), Play(value=0, description='play/pause', interval=10, max=99), BoundedIntText(value=1, description='Step Size:', layout=Layout(width='150px'), max=99, min=1, tooltip='set slider step'), BoundedIntText(value=100, description='Speed', layout=Layout(width='150px'), max=1000, min=1, step=50, tooltip='Playback speed, this is NOT framerate.\\nArbitrary units between 1 - 1,000'))), IntSlider(value=0, description='dimension: t', max=99)))))"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iw_zfish.show(sidecar=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b4c57f1-d2a8-4607-b515-7c733c3c8f23",
   "metadata": {},
   "source": [
    "### Apply a window function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "f595ddf0-5dfd-486e-ac4c-b34aa64862a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "iw_zfish.window_funcs[\"t\"].func = np.max"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "cb60d702-db37-42bf-9ada-298006ae46df",
   "metadata": {},
   "outputs": [],
   "source": [
    "# close the plot\n",
    "iw_zfish.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb90e4bc-37e4-4f56-8350-f821f26c8524",
   "metadata": {},
   "source": [
    "### z-slider\n",
    "\n",
    "**ImageWidget will also give you a slider for \"z\" in addition to \"t\" if necessary**\n",
    "\n",
    "This example uses the same example data shown above, but displays them in a single subplot and ImageWidget provides a z-slider. You can use window_funcs, frame_apply funcs, etc. There is no difference in ImageWidget behavior with the z-slider."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "13299d8d-2748-476c-a8b9-69c05321ee44",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0c3bb5def3ba4500b0ba3c2c70a4eec0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "RFBOutputContext()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "iw_z = fpl.ImageWidget(\n",
    "    data=zfish_data, # you can also provide a list of tzxy arrays\n",
    "    window_funcs={\"t\": (np.mean, 5)},\n",
    "    cmap=\"gnuplot2\", \n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "201bf7c3-2d36-4fe6-9f7b-5dc5f1afe1ef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "JupyterOutputContext(children=(JupyterWgpuCanvas(), IpywidgetToolBar(children=(Button(icon='expand-arrows-alt'…"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iw_z.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "a7db958b-623f-4157-926c-d5045d9d0116",
   "metadata": {},
   "outputs": [],
   "source": [
    "# close the plot\n",
    "iw_z.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bcf9d9b2-7a3e-4680-80d9-79d8f40138a8",
   "metadata": {},
   "source": [
    "## 2D Lines Practice"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "7f5a6b74-693a-4cb6-bec5-d8464b0a9da6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6fe8b4f0583e41c1a426b6d1cbb3cc76",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "RFBOutputContext()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "JupyterOutputContext(children=(JupyterWgpuCanvas(), IpywidgetToolBar(children=(Button(icon='expand-arrows-alt', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='auto-scale scene'), Button(icon='align-center', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='auto-center scene'), ToggleButton(value=True, icon='hand-pointer', layout=Layout(width='auto'), tooltip='panzoom controller'), ToggleButton(value=True, description='1:1', layout=Layout(width='auto'), style=ToggleButtonStyle(font_weight='bold'), tooltip='maintain aspect'), Button(icon='arrow-up', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='y-axis direction'), Button(icon='draw-polygon', layout=Layout(width='auto'), style=ButtonStyle(), tooltip='add PolygonSelector'), ToggleButton(value=False, icon='video', layout=Layout(width='auto'), tooltip='record'), Dropdown(description='Subplots:', layout=Layout(width='200px'), options=('(0, 0)',), value='(0, 0)')))))"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# same plot with sine and cosine wave only\n",
    "# Create a figure\n",
    "fig_lines = fpl.Figure()\n",
    "\n",
    "# we will add all the lines to the same subplot\n",
    "subplot = fig_lines[0, 0]\n",
    "\n",
    "# plot sine wave, use a single color\n",
    "sine_graphic = subplot.add_line(data=sine, thickness=5, colors=\"magenta\")\n",
    "\n",
    "# you can also use colormaps for lines!\n",
    "cosine_graphic = subplot.add_line(data=cosine, thickness=12, cmap=\"autumn\", offset=(0, 5, 0))\n",
    "\n",
    "# show the plot\n",
    "fig_lines.show(sidecar=True, sidecar_kwargs={\"title\": \"lines\"})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35e30550-6ed7-445e-b944-48b70fb5299e",
   "metadata": {},
   "source": [
    "**Question:** Can you change the colormap of the sine_graphic to \"hsv\"?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5dc7f467-51e7-4df4-9856-8c897a407d68",
   "metadata": {},
   "outputs": [],
   "source": [
    "### insert code here\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d3666323-850c-4768-8b6c-ea0c2ee609aa",
   "metadata": {},
   "source": [
    "**Question:** Can you change the color of first 50 data points of the cosine_graphic to green?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ad6963da-bd09-49ca-84c2-0fc94d873cd2",
   "metadata": {},
   "outputs": [],
   "source": [
    "### insert code here\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "813905b1-1529-4df1-a5b8-8f8b2d1372a3",
   "metadata": {},
   "source": [
    "**Question:** Can you to change the color of last 50 data points of the cosine_graphic to equal the colors of the last 50 data points of the sine_graphic?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1d62006a-4e6c-402d-8813-b8cf8cf1739d",
   "metadata": {},
   "outputs": [],
   "source": [
    "### insert code here\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91e4cfce-a8e7-45f5-9221-3542fdfc93bc",
   "metadata": {},
   "source": [
    "#### Capture changes to graphic properties as events"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7c141283-c2ea-4166-a4f5-b36d83c846ee",
   "metadata": {},
   "source": [
    "**Two ways to add events in `fastplotlib`**\n",
    "\n",
    "1) Using `graphic.add_event_handler(callback_func, \"property\")`\n",
    "\n",
    "\n",
    "2) Using a decorator:\n",
    "```\n",
    "@graphic.add_event_handler(\"property\")\n",
    "def callback_func(ev):\n",
    "    pass\n",
    "```\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "233ef1cf-55a4-4da5-a85a-57afcde12221",
   "metadata": {},
   "source": [
    "#### Examples:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "ae2abe38-83c6-4505-9734-293ec727405e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# first set the color of the sine and cosine wave to white so we can better see the changes\n",
    "cosine_graphic.colors = \"w\"\n",
    "sine_graphic.colors = \"w\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "03c0a1d7-5504-4ad6-a857-1a94f62f5934",
   "metadata": {},
   "outputs": [],
   "source": [
    "# will print event data when the color of the cosine graphic changes\n",
    "@cosine_graphic.add_event_handler(\"colors\")\n",
    "def callback_func(ev):\n",
    "    print(ev.info)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "3b883e39-ecd3-4e69-a7fd-52bd628a6061",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<function __main__.change_colors(ev)>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# when the cosine graphic colors change, will also update the sine_graphic colors\n",
    "def change_colors(ev):\n",
    "   sine_graphic.colors[ev.info[\"key\"]] = \"pink\"\n",
    "\n",
    "cosine_graphic.add_event_handler(change_colors, \"colors\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "81b89bc8-0056-4fe1-bf28-455ed26f41c8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'key': slice(None, 10, None), 'value': array([[1., 0., 0., 1.],\n",
      "       [1., 0., 0., 1.],\n",
      "       [1., 0., 0., 1.],\n",
      "       [1., 0., 0., 1.],\n",
      "       [1., 0., 0., 1.],\n",
      "       [1., 0., 0., 1.],\n",
      "       [1., 0., 0., 1.],\n",
      "       [1., 0., 0., 1.],\n",
      "       [1., 0., 0., 1.],\n",
      "       [1., 0., 0., 1.]], dtype=float32), 'user_value': 'r'}\n"
     ]
    }
   ],
   "source": [
    "cosine_graphic.colors[:10] = \"r\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "19b07476-8687-410b-bdd8-42075952837a",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig_lines.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ab5f1d0a-085b-44ab-ad77-058dd3ed9c74",
   "metadata": {},
   "source": [
    "## More Events :D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "6a05999b-e3ba-4935-aaa0-20c678bfa0bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "# generate some circles\n",
    "def make_circle(center, radius: float, n_points: int = 75) -> np.ndarray:\n",
    "    theta = np.linspace(0, 2 * np.pi, n_points)\n",
    "    xs = radius * np.sin(theta)\n",
    "    ys = radius * np.cos(theta)\n",
    "\n",
    "    return np.column_stack([xs, ys]) + center\n",
    "\n",
    "# this makes 5 circles, so we can create 5 cmap values, so it will use these values to set the\n",
    "# color of the line based by using the cmap as a LUT with the corresponding cmap_value\n",
    "circles = list()\n",
    "for x in range(0, 50, 10):\n",
    "    circles.append(make_circle(center=(x, 0), radius=4, n_points=100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "e51846f3-8a3e-41c7-be1b-ae9f4e63a080",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d81c04254d8f46d18f6afbd97ed3e70d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "RFBOutputContext()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = fpl.Figure()\n",
    "\n",
    "circles_graphic = fig[0,0].add_line_collection(data=circles, cmap=\"tab10\", thickness=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "4d06240c-2471-4c18-be2f-6d11fdfb044f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d7907ee50af14a3cb0ec30436b24c2a3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "JupyterOutputContext(children=(JupyterWgpuCanvas(), IpywidgetToolBar(children=(Button(icon='expand-arrows-alt'…"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "498a31e6-a49f-479e-8955-05ee9ac3a3ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "# get the nearest graphic that is clicked and change the color\n",
    "def click_event(ev):\n",
    "    # reset colors\n",
    "    circles_graphic.cmap = \"tab10\"\n",
    "\n",
    "    # map the click position to world coordinates\n",
    "    xy = fig[0, 0].map_screen_to_world(ev)[:-1]\n",
    "\n",
    "    # get the nearest graphic to the position\n",
    "    nearest = fpl.utils.get_nearest_graphics(xy, circles_graphic)[0]\n",
    "\n",
    "    # change the closest graphic color to white\n",
    "    nearest.colors = \"w\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "715dd0a3-3b34-41b4-b2c1-8990151e8170",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<function __main__.click_event(ev)>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# add event handler to the renderer\n",
    "fig[0,0].renderer.add_event_handler(click_event, \"click\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d3fe2917-89c3-4fc2-be8e-bdbda6229547",
   "metadata": {},
   "source": [
    "### Events Practice\n",
    "\n",
    "**Note:** If you make a mistake adding the event handler, you can easily clear the event handlers using the following notation:\n",
    "```\n",
    "circles_graphic.clear_event_handlers()\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8954927-bb2f-47d5-92b9-22af581a9a2f",
   "metadata": {},
   "source": [
    "**Question:** Can you add another event handler (using either method) to the circles_graphic that will change the thickness of a circle to 5 when you hover over it?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d628720c-94ec-4ea8-be63-16746ebd127a",
   "metadata": {},
   "outputs": [],
   "source": [
    "### insert code here\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e9b9ebed-cff1-40a5-b20d-1b6b2cb21d5d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# close the plot\n",
    "fig.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fd73d41-9d9d-4c0e-adab-99d6cfb0db7b",
   "metadata": {},
   "source": [
    "## Selectors"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f64c884-2896-4fd4-99d4-f26f6caa104b",
   "metadata": {},
   "source": [
    "#### Let's look at a `LinearSelector`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "a9114d6d-d472-4140-93ba-71b53511cbea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "44f2a75cb97c461eaf13fcfcd8114657",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "RFBOutputContext()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c94b235d9ff7492586a82d3ad9ff6f4e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "JupyterOutputContext(children=(JupyterWgpuCanvas(), IpywidgetToolBar(children=(Button(icon='expand-arrows-alt'…"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fig = fpl.Figure()\n",
    "\n",
    "# same sine data from before\n",
    "sine_graphic = fig[0, 0].add_line(data=sine, colors=\"w\")\n",
    "\n",
    "# add a linear selector the sine wave\n",
    "selector = sine_graphic.add_linear_selector()\n",
    "\n",
    "\n",
    "fig[0, 0].auto_scale()\n",
    "fig.show(maintain_aspect=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "58df2108-c4c6-49a5-9ed5-6190145b8f96",
   "metadata": {},
   "outputs": [],
   "source": [
    "# change the color of the sine wave based on the location of the linear selector\n",
    "@selector.add_event_handler(\"selection\")\n",
    "def set_color_at_index(ev):\n",
    "    # get the selected index\n",
    "    ix = ev.get_selected_index()\n",
    "    # get the sine graphic\n",
    "    g = ev.graphic.parent\n",
    "    # change the color of the sine graphic at the index of the selector\n",
    "    g.colors[ix] = \"green\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "18e2344d-c608-4079-80c9-0e482514818f",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21ed3b48-1d8b-47d2-9bf2-7543c64f9c7c",
   "metadata": {},
   "source": [
    "#### Now let's look at a `LinearRegionSelector`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "c440b95a-7d76-46ac-aa52-966bb9648a6e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3143c785457349178fd8d52876c9fb57",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "RFBOutputContext()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9ef3d4fb18f34adf9a8a8599bcb807bd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "JupyterOutputContext(children=(JupyterWgpuCanvas(), IpywidgetToolBar(children=(Button(icon='expand-arrows-alt'…"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fig = fpl.Figure((2, 1))\n",
    "\n",
    "# data to plot\n",
    "xs = np.linspace(0, 10* np.pi, 1_000)\n",
    "sine = np.sin(xs)\n",
    "sine += 100\n",
    "\n",
    "# make sine along x axis\n",
    "sine_graphic_x = fig[0, 0].add_line(np.column_stack([xs, sine]), offset=(10, 0, 0))\n",
    "\n",
    "# add a linear region selector\n",
    "ls_x = sine_graphic_x.add_linear_region_selector()  # default axis is \"x\"\n",
    "\n",
    "# get the initial selected date of the linear region selector\n",
    "zoomed_init = ls_x.get_selected_data()\n",
    "\n",
    "# make a line graphic for displaying zoomed data\n",
    "zoomed_x = fig[1, 0].add_line(zoomed_init)\n",
    "\n",
    "@ls_x.add_event_handler(\"selection\")\n",
    "def set_zoom_x(ev):\n",
    "    \"\"\"sets zoomed x selector data\"\"\"\n",
    "    # get the selected data\n",
    "    selected_data = ev.get_selected_data()\n",
    "\n",
    "    # remove the current zoomed data\n",
    "    # and update with new selected data\n",
    "    global zoomed_x\n",
    "    fig[1,0].remove_graphic(zoomed_x)\n",
    "    zoomed_x = fig[1,0].add_line(selected_data)\n",
    "    fig[1, 0].auto_scale()\n",
    "\n",
    "fig.show(maintain_aspect=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8b454250-eb4f-4f71-a743-7432886907c9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# close the plot\n",
    "fig.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "475f97f8-c5a2-4c02-9469-32543a7f7827",
   "metadata": {},
   "source": [
    "## For more examples, please see our gallery :D"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fc949dee-1d31-4e8b-9548-414ade39f7df",
   "metadata": {},
   "source": [
    "### https://fastplotlib.readthedocs.io/en/latest/_gallery/index.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "316cb8b2-7708-41b2-bc6b-d4720629c5bd",
   "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": "3.11.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}