love_multiple.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import networkx as nx\n",
    "import random\n",
    "import numpy as np\n",
    "import ipywidgets as widgets\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.pyplot import show\n",
    "from ipywidgets import GridBox, Layout, interactive_output\n",
    "import love.love_multiple as multiple\n",
    "from IPython.display import clear_output\n",
    "from IPython.display import display\n",
    "import pickle\n",
    "random.seed(0)\n",
    "num = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# network parmaeters and output\n",
    "number_of_nodes = 10\n",
    "G = nx.complete_graph(number_of_nodes)\n",
    "nx.set_edge_attributes(G, [], \"t\")\n",
    "nx.set_edge_attributes(G, [], \"data\")\n",
    "nx.set_edge_attributes(G, [], \"parameters\")\n",
    "out = widgets.Output()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#initialize each node with the romantic parameters (matrix and vector) for a different node\n",
    "G = multiple.randomize_parameter_network(G, number_of_nodes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "slider_style = {'description_width': 'initial'} \n",
    "slider_layout = Layout(width='auto')\n",
    "# Parameters\n",
    "\n",
    "#initial starting parameters for graph\n",
    "# Matrix A\n",
    "# G[0][1]['parameters'][0][0][0] means select edge from node 0 to node 1, select the parameter attribute, matrix A, row 0, column 0\n",
    "axx= G[0][1]['parameters'][0][0][0]; axy= G[0][1]['parameters'][0][0][1]; bxx= G[0][1]['parameters'][0][0][2]; bxy= G[0][1]['parameters'][0][0][3]\n",
    "ayx= G[0][1]['parameters'][0][1][0]; ayy= G[0][1]['parameters'][0][1][1]; byx= G[0][1]['parameters'][0][1][2]; byy= G[0][1]['parameters'][0][1][3]\n",
    "cxx= G[0][1]['parameters'][0][2][0]; cxy= G[0][1]['parameters'][0][2][1]; dxx= G[0][1]['parameters'][0][2][2]; dxy= G[0][1]['parameters'][0][2][3]\n",
    "cyx= G[0][1]['parameters'][0][3][0]; cyy= G[0][1]['parameters'][0][3][1]; dyx= G[0][1]['parameters'][0][3][2]; dyy= G[0][1]['parameters'][0][3][3]\n",
    "#Vector B\n",
    "# G[0][1]['parameters'][1][0] means select eedge from node 0 to node 1, select the parameter attirbute, vector B, row 0\n",
    "fxy=G[0][1]['parameters'][1][0]; fyx=G[0][1]['parameters'][1][1]; gxy=G[0][1]['parameters'][1][2]; gyx=G[0][1]['parameters'][1][3]\n",
    "# Vector C\n",
    "# G[0][1]['parameters'][2][0] means select edge form node 0 to node 1, select the parameter attribute, vector C, row 0\n",
    "xi0=G[0][1]['parameters'][2][0]; yi0=G[0][1]['parameters'][2][1]; xp0=G[0][1]['parameters'][2][2]; yp0=G[0][1]['parameters'][2][3]\n",
    "\n",
    "#dropdown selectors\n",
    "node1 = widgets.Dropdown(options=[str(i) for i in range(number_of_nodes)], value='0', description='Person:')\n",
    "node2 = widgets.Dropdown(options=[str(i) for i in range(number_of_nodes)], value='1', description='Person:') \n",
    "\n",
    "# sliders to update parameters for selected people\n",
    "axx_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=axx, description='axx', style=slider_style, layout=slider_layout)\n",
    "axy_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=axy, description='axy', style=slider_style, layout=slider_layout)\n",
    "bxx_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=bxx, description='bxx', style=slider_style, layout=slider_layout)\n",
    "bxy_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=bxy, description='bxy', style=slider_style, layout=slider_layout)\n",
    "ayx_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=ayx, description='ayx', style=slider_style, layout=slider_layout)\n",
    "ayy_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=ayy, description='ayy', style=slider_style, layout=slider_layout)\n",
    "byx_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=byx, description='byx', style=slider_style, layout=slider_layout)\n",
    "byy_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=byy, description='byy', style=slider_style, layout=slider_layout)\n",
    "cxx_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=cxx, description='cxx', style=slider_style, layout=slider_layout)\n",
    "cxy_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=cxy, description='cxy', style=slider_style, layout=slider_layout)\n",
    "dxx_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=dxx, description='dxx', style=slider_style, layout=slider_layout)\n",
    "dxy_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=dxy, description='dxy', style=slider_style, layout=slider_layout)\n",
    "cyx_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=cyx, description='cyx', style=slider_style, layout=slider_layout)\n",
    "cyy_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=cyy, description='cyy', style=slider_style, layout=slider_layout)\n",
    "dyx_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=dyx, description='dyx', style=slider_style, layout=slider_layout)\n",
    "dyy_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=dyy, description='dyy', style=slider_style, layout=slider_layout)\n",
    "\n",
    "fxy_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=fxy, description='fxy', style=slider_style, layout=slider_layout)\n",
    "fyx_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=fyx, description='fyx', style=slider_style, layout=slider_layout)\n",
    "gxy_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=gxy, description='gxy', style=slider_style, layout=slider_layout)\n",
    "gyx_slider = widgets.FloatSlider(min=-1.0, max=1.0, step=0.01, value=gyx, description='gyx', style=slider_style, layout=slider_layout)\n",
    "\n",
    "xi0_slider = widgets.FloatSlider(min=-1_000_000_000, max=1_000_000_000, step=1, value=xi0, description='xi0', style=slider_style, layout=slider_layout)\n",
    "yi0_slider = widgets.FloatSlider(min=-1_000_000_000, max=1_000_000_000, step=1, value=yi0, description='yi0', style=slider_style, layout=slider_layout)\n",
    "xp0_slider = widgets.FloatSlider(min=-1_000_000_000, max=1_000_000_000, step=1, value=xp0, description='xp0', style=slider_style, layout=slider_layout)\n",
    "yp0_slider = widgets.FloatSlider(min=-1_000_000_000, max=1_000_000_000, step=1, value=yp0, description='yp0', style=slider_style, layout=slider_layout)\n",
    "\n",
    "#buttons to interact with\n",
    "savebutton = widgets.Button(description=\"Save Parameters\", layout=Layout(width='200px'))\n",
    "randomizebutton = widgets.Button(description=\"Randomize These Parameters\", layout=Layout(width='200px'))\n",
    "randomizeallbutton = widgets.Button(description=\"Randomize All Parameters\", layout=Layout(width='200px'))\n",
    "updateplotbutton = widgets.Button(description=\"Update Plot (Not Saved)\", layout=Layout(width='200px'))\n",
    "downloadGbutton = widgets.Button(description=\"Save Graph To File\", layout=Layout(width='200px'))\n",
    "uploadGfilebutton = widgets.Button(description=\"Click To Upload File (.pickle File)\", layout=Layout(width='400px'))\n",
    "\n",
    "# text area to insert file name and to output status messages\n",
    "uploadGfile = widgets.Text(value='', placeholder='Give the path of the file you want to upload', layout=Layout(width='1000px'))\n",
    "statustext = widgets.Text(value='', placeholder='Message/console output', disabled=True, layout=Layout(width='1000px'))\n",
    "\n",
    "\n",
    "selectpeople = widgets.SelectMultiple(options=[i for i in range(number_of_nodes)],value=[1, 2], description='People Selector For Top Right Graph', layout=Layout(width='200px'))\n",
    "\n",
    "# checkbox to normalzie vlaues\n",
    "normalizecheckbox = widgets.Checkbox(value=False,description='Normalize values across all connections with sum=0 when saving/randomizing parameters', layout=Layout(width='800px'))\n",
    "normalize1checkbox = widgets.Checkbox(value=False,description='Scale absolute value of normalized values to 1 when saving/randomizing parameters', layout=Layout(width='800px'))\n",
    "\n",
    "def update_slider_values():\n",
    "    \"\"\"When needed, call this function to update the slider values with the values of the edge defined betweeen the nodes selected in the dropdowns. \n",
    "    Doesn't take any arguments or returns anything, updates the global slider values\n",
    "    \"\"\"\n",
    "    node1_value = int(node1.value)\n",
    "    node2_value = int(node2.value)\n",
    "    axx_slider.value = G[node1_value][node2_value]['parameters'][0][0][0]\n",
    "    axy_slider.value = G[node1_value][node2_value]['parameters'][0][0][1]\n",
    "    bxx_slider.value = G[node1_value][node2_value]['parameters'][0][0][2]\n",
    "    bxy_slider.value = G[node1_value][node2_value]['parameters'][0][0][3]\n",
    "    ayx_slider.value = G[node1_value][node2_value]['parameters'][0][1][0]\n",
    "    ayy_slider.value = G[node1_value][node2_value]['parameters'][0][1][1]\n",
    "    byx_slider.value = G[node1_value][node2_value]['parameters'][0][1][2]\n",
    "    byy_slider.value = G[node1_value][node2_value]['parameters'][0][1][3]\n",
    "    cxx_slider.value = G[node1_value][node2_value]['parameters'][0][2][0]\n",
    "    cxy_slider.value = G[node1_value][node2_value]['parameters'][0][2][1]\n",
    "    dxx_slider.value = G[node1_value][node2_value]['parameters'][0][2][2]\n",
    "    dxy_slider.value = G[node1_value][node2_value]['parameters'][0][2][3]\n",
    "    cyx_slider.value = G[node1_value][node2_value]['parameters'][0][3][0]\n",
    "    cyy_slider.value = G[node1_value][node2_value]['parameters'][0][3][1]\n",
    "    dyx_slider.value = G[node1_value][node2_value]['parameters'][0][3][2]\n",
    "    dyy_slider.value = G[node1_value][node2_value]['parameters'][0][3][3]\n",
    "\n",
    "    fxy_slider.value = G[node1_value][node2_value]['parameters'][1][0]\n",
    "    fyx_slider.value = G[node1_value][node2_value]['parameters'][1][1]\n",
    "    gxy_slider.value = G[node1_value][node2_value]['parameters'][1][2]\n",
    "    gyx_slider.value = G[node1_value][node2_value]['parameters'][1][3]\n",
    "\n",
    "    xi0_slider.value = G[node1_value][node2_value]['parameters'][2][0]\n",
    "    yi0_slider.value = G[node1_value][node2_value]['parameters'][2][1]\n",
    "    xp0_slider.value = G[node1_value][node2_value]['parameters'][2][2]\n",
    "    yp0_slider.value = G[node1_value][node2_value]['parameters'][2][3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def update_plot(t_data = None, y_data = None):\n",
    "    \"\"\"Updates the plot and plots the new data. Call this whenever you want to update the plot\n",
    "\n",
    "    Args:\n",
    "        t_data (list, optional): pass selected data and graph that. Data passed is the time series data of the return from solve_ivp. Defaults to None.\n",
    "        y_data (list, optional): pass selected data and graph that. Data passed is the y series data of the return from solve_ivp. Defaults to None.\n",
    "    \"\"\"\n",
    "    with out:\n",
    "        clear_output()\n",
    "        node1_value = int(node1.value)\n",
    "        node2_value = int(node2.value)\n",
    "        if (node1_value == node2_value):\n",
    "            statustext.value = \"Error: Please select two different people\"\n",
    "            return\n",
    "        if (t_data is not None and y_data is not None):\n",
    "            t = t_data\n",
    "            xa = y_data\n",
    "        else:\n",
    "            xa = G[int(node1_value)][int(node2_value)][\"data\"]\n",
    "            t = G[int(node1_value)][int(node2_value)][\"t\"]\n",
    "        \n",
    "        # Set up a figure twice as tall as it is wide\n",
    "        fig = plt.figure(figsize=(10, 10))\n",
    "        # first subplot\n",
    "        ax = fig.add_subplot(2, 2, 1)\n",
    "        ax.scatter(xa[0], xa[1], s=50, cmap='viridis', alpha=0.5, c=t)\n",
    "        ax.scatter(xa[2], xa[3], s=50, cmap='viridis', alpha=0.5, c=t)\n",
    "        ax.plot(xa[0],xa[1], c=\"black\", label=\"Intimacy\", alpha=0.5)\n",
    "        ax.plot(xa[2],xa[3], c=\"red\", label=\"Passion\", alpha=0.5)\n",
    "        ax.set_xlabel(\"Person \" + node1.value, fontsize=10)\n",
    "        ax.set_ylabel(\"Person \" + node2.value, fontsize=10)\n",
    "        ax.grid(True)\n",
    "        ax.legend()\n",
    "\n",
    "        # second subplot\n",
    "        ax = fig.add_subplot(2, 2, 2)\n",
    "        ax.scatter(xa[0], xa[2], s=50, cmap='viridis', alpha=0.5, c=t)\n",
    "        ax.scatter(xa[1], xa[3], s=50, cmap='viridis', alpha=0.5, c=t)\n",
    "        ax.plot(xa[0],xa[2], c=\"black\", label=\"Person \" + node2.value, alpha=0.5)\n",
    "        ax.plot(xa[1],xa[3], c=\"red\", label=\"Person \" + node1.value, alpha=0.5)\n",
    "        ax.set_xlabel('Intimacy', fontsize=10)\n",
    "        ax.set_ylabel('Passion', fontsize=10)\n",
    "        ax.grid(True)\n",
    "        ax.legend()\n",
    "\n",
    "        # third subplot\n",
    "        ax = fig.add_subplot(2, 2, 3, projection='3d')\n",
    "        ax.scatter(xa[0], xa[1], xa[2], c=t, s=30, cmap='viridis', alpha=0.5)\n",
    "        x_lim = ax.get_xlim()\n",
    "        y_lim = ax.get_ylim()\n",
    "        z_lim = ax.get_zlim()\n",
    "        ax.plot(xa[0], xa[2], 'r+', zdir='y', zs=y_lim[1], alpha=0.5)\n",
    "        ax.plot(xa[1], xa[2], 'g+', zdir='x', zs=x_lim[0], alpha=0.5)\n",
    "        ax.plot(xa[0], xa[1], 'b+', zdir='z', zs=z_lim[0], alpha=0.5)\n",
    "        ax.set_xlabel('Intimacy of Person ' + node1.value, fontsize=10)\n",
    "        ax.set_ylabel('Intimacy of Person ' + node2.value, fontsize=10)\n",
    "        ax.set_zlabel('Passion of Person ' + node1.value, fontsize=10)\n",
    "        ax.grid(True)\n",
    "\n",
    "        # fourth subplot\n",
    "        ax = fig.add_subplot(2, 2, 4, projection='3d')\n",
    "        ax.scatter(xa[0], xa[1], xa[3], c=t, cmap='viridis', alpha=0.5, s=30)\n",
    "        x_lim = ax.get_xlim()\n",
    "        y_lim = ax.get_ylim()\n",
    "        z_lim = ax.get_zlim()\n",
    "        ax.plot(xa[0], xa[3], 'r+', zdir='y', zs=y_lim[1], alpha=0.5)\n",
    "        ax.plot(xa[1], xa[3], 'g+', zdir='x', zs=x_lim[0], alpha=0.5)\n",
    "        ax.plot(xa[0], xa[1], 'b+', zdir='z', zs=z_lim[0], alpha=0.5)\n",
    "        ax.set_xlabel('Intimacy of Person ' + node1.value, fontsize=10)\n",
    "        ax.set_ylabel('Intimacy of Person ' + node2.value, fontsize=10)\n",
    "        ax.set_zlabel('Passion of Person ' + node2.value, fontsize=10)\n",
    "        ax.grid(True)\n",
    "        plt.show()\n",
    "        show()\n",
    "\n",
    "def normalize_values():\n",
    "    \"\"\"A fully connected network of size n has n(n-1)/2 edges, consisting of a matrix A, vector B and C. Loop through each individually, grab each element from each position element-wise, normalize the values, and put them bakc in their respective positions. \n",
    "    For example witjh matrix [[1, 2], [3, 4]] and matrix [[5, 6], [7, 8]], (it is posisble to have m matrices of size n) the program takes [1, 5], [2, 6], [3, 7], [4, 8] (ithe length of these matrices would then be m) normalizes the individual vectors, and returns the values into their respective vectors. [1/sqrt(26), 25/sqrt(26)], [1/sqrt(10), 3/sqrt(10)], [3/sqrt(58), 7/sqrt(58)], [1/sqrt(5), 2/sqrt(5)]]. The final matrix would then be [[1/sqrt(26), 1/sqrt(10)], [3/sqrt(58), 1/sqrt(5)]] and [[25/sqrt(26), 3/sqrt(10)], [7/sqrt(58), 2/sqrt(5)]]. Does this for all possible edges. \n",
    "    Modifies G by inserting the values back into their respective edges\n",
    "    \"\"\"\n",
    "    matrixA = [\n",
    "        [[], [], [], []], \n",
    "        [[], [], [], []], \n",
    "        [[], [], [], []], \n",
    "        [[], [], [], []]\n",
    "    ]\n",
    "    vectorB = [[], [], [], []]\n",
    "    vectorC = [[], [], [], []]\n",
    "    for i, j, d in G.edges(data=True):\n",
    "        a = d['parameters'][0]\n",
    "        b = d['parameters'][1]\n",
    "        c = d['parameters'][2]\n",
    "        for row in range(len(a)):\n",
    "            for col in range(len(a[0])):\n",
    "                matrixA[row][col].append(a[row][col])\n",
    "        for row in range(len(b)):\n",
    "            vectorB[row].append(b[row])\n",
    "        for row in range(len(c)):\n",
    "            vectorC[row].append(b[row])\n",
    "\n",
    "    for row in range(len(matrixA)):\n",
    "            for col in range(len(matrixA[0])):\n",
    "                norm = np.array(matrixA[row][col])/np.linalg.norm(matrixA[row][col])\n",
    "                final_vector = norm - np.mean(norm)\n",
    "                if normalize1checkbox.value:\n",
    "                    scaling_factor = np.max(np.abs(final_vector))\n",
    "                    final_vector = final_vector / scaling_factor\n",
    "                matrixA[row][col] = final_vector\n",
    "    vectorBp = [[], [], [], []]\n",
    "    for row in range(len(vectorB)):\n",
    "        for i in range(len(vectorB[row])):\n",
    "            vectorBp[row].append(vectorB[row][i])\n",
    "    for row in range(len(vectorBp)):\n",
    "        norm = np.array(vectorBp[row])/np.linalg.norm(vectorBp[row])\n",
    "        final_vector = norm - np.mean(norm)\n",
    "        if normalize1checkbox.value:\n",
    "            scaling_factor = np.max(np.abs(final_vector))\n",
    "            final_vector = final_vector / scaling_factor\n",
    "        vectorBp[row] = final_vector\n",
    "    vectorCp = [[], [], [], []]\n",
    "    for row in range(len(vectorC)):\n",
    "        for i in range(len(vectorC[row])):\n",
    "            vectorCp[row].append(vectorC[row][i])\n",
    "    for row in range(len(vectorCp)):\n",
    "        norm = np.array(vectorCp[row])/np.linalg.norm(vectorCp[row])\n",
    "        final_vector = norm - np.mean(norm)\n",
    "        if normalize1checkbox.value:\n",
    "            scaling_factor = np.max(np.abs(final_vector))\n",
    "            final_vector = final_vector / scaling_factor\n",
    "        vectorCp[row] = final_vector\n",
    "    \n",
    "    counter = 0\n",
    "    for i, j, d in G.edges(data=True):\n",
    "        matrixa = np.array([[matrixA[0][0][counter], matrixA[0][1][counter], matrixA[0][2][counter], matrixA[0][3][counter]], [matrixA[1][0][counter], matrixA[1][1][counter], matrixA[1][2][counter], matrixA[1][3][counter]], [matrixA[2][0][counter], matrixA[2][1][counter], matrixA[2][2][counter], matrixA[2][3][counter]], [matrixA[3][0][counter], matrixA[3][1][counter], matrixA[3][2][counter], matrixA[3][3][counter]]])\n",
    "        G[i][j]['parameters'][0] = matrixa\n",
    "\n",
    "        vectorb = np.array([vectorBp[0][counter], vectorBp[1][counter], vectorBp[2][counter], vectorBp[3][counter]])\n",
    "        G[i][j]['parameters'][1] = vectorb\n",
    "        \n",
    "        vectorc = np.array([vectorCp[0][counter], vectorCp[1][counter], vectorCp[2][counter], vectorCp[3][counter]])\n",
    "        G[i][j]['parameters'][2] = vectorc\n",
    "        counter += 1\n",
    "\n",
    "def save_graph(change):\n",
    "    \"\"\"saves the current graph object into a pickle file and notifies user that file has been saved to storage\n",
    "\n",
    "    Args:\n",
    "        change (array): automatically generated by ipywidgets\n",
    "    \"\"\"\n",
    "    pickle.dump(G, open('love_multiple.pickle', 'wb'))\n",
    "    statustext.value = \"Success: saved graph data to file with filename \\\"love_multiple.pickle\\\"\"\n",
    "\n",
    "# load graph object from file\n",
    "def upload_graph(change):\n",
    "    \"\"\"uploads graph to the network and overwrites the current network. \n",
    "    To actually upload a file, insert full or local pathname to file\n",
    "\n",
    "    Args:\n",
    "        change (array): automatically generated by ipywidgets\n",
    "    \"\"\"\n",
    "    global G\n",
    "    if ('.pickle' not in uploadGfile.value):\n",
    "        statustext.value = \"Error: Invalid file path. Please enter a .pickle file\"\n",
    "        return\n",
    "    if (uploadGfile.value == ''):\n",
    "        statustext.value = \"Please insert a path to the filename (either full or relative)\"\n",
    "        return\n",
    "    try:\n",
    "        tempG = pickle.load(open(uploadGfile.value, 'rb'))\n",
    "    except:\n",
    "        statustext.value = \"Error: Problem loading file. Try again\"\n",
    "        return\n",
    "    G = tempG\n",
    "    statustext.value = \"Success: loaded graph data from file into graph object\"\n",
    "    update_slider_values()\n",
    "    update_plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# gorup up the buttons and sliders end more\n",
    "dropdown = [node1, node2] \n",
    "button = [savebutton, randomizebutton, randomizeallbutton, updateplotbutton, downloadGbutton]\n",
    "select= [selectpeople]\n",
    "checkbox = [normalizecheckbox, normalize1checkbox, uploadGfile, uploadGfilebutton, statustext]\n",
    "sliders = [axx_slider,axy_slider,bxx_slider,bxy_slider,ayx_slider,ayy_slider,byx_slider,byy_slider, cxx_slider,cxy_slider,dxx_slider,dxy_slider,cyx_slider,cyy_slider,dyx_slider,dyy_slider,fxy_slider,fyx_slider,gxy_slider,gyx_slider,xi0_slider,yi0_slider,xp0_slider,yp0_slider]\n",
    "\n",
    "def update_sliders(change):\n",
    "    \"\"\"Dropdown to select the new nodes\n",
    "\n",
    "    Args:\n",
    "        change (array): automatically called by ipywidgets\n",
    "    \"\"\"\n",
    "    node1_value = int(node1.value)\n",
    "    node2_value = int(node2.value)\n",
    "    if (node1_value == node2_value):\n",
    "        statustext.value = \"Error: Select two different people\"\n",
    "        return\n",
    "    if (node1_value > node2_value):\n",
    "        temp = node1_value\n",
    "        node1.value = node2.value\n",
    "        node2.value = temp\n",
    "        statustext.value = \"Success: updated sliders to reflect parameters for person (but the values have been swapped)\" + node1.value + \" and person \" + node2.value\n",
    "    else:\n",
    "        statustext.value = \"Success: updated sliders to reflect parameters for person \" + node1.value + \" and person \" + node2.value\n",
    "    \n",
    "    update_slider_values()\n",
    "    update_plot()\n",
    "\n",
    "def update_graph2(change):\n",
    "    \"\"\"update plot\"\"\"\n",
    "    update_plot()\n",
    "\n",
    "# observe certain objects for change\n",
    "node1.observe(update_sliders, names='value')\n",
    "node2.observe(update_sliders, names='value')\n",
    "selectpeople.observe(update_graph2, names='value')\n",
    "\n",
    "# layouting\n",
    "grid_dropdown = GridBox(dropdown, layout=Layout(\n",
    "    width='100%',\n",
    "    grid_template_columns='repeat(2, 30%)',  \n",
    "    grid_gap='20px 20px'\n",
    "))\n",
    "grid_buttons = GridBox(button, layout=Layout(\n",
    "    width='100%',\n",
    "    grid_template_columns='repeat(5, 15%)',  \n",
    "    grid_gap='20px 20px'\n",
    "))\n",
    "# grid_select = GridBox(select, layout=Layout(\n",
    "#     width='100%',\n",
    "#     grid_template_columns='repeat(5, 15%)',  \n",
    "#     grid_gap='20px 20px'\n",
    "# ))\n",
    "grid_checkboxes = GridBox(checkbox, layout=Layout(\n",
    "    width='100%',\n",
    "    grid_template_columns='repeat(1, 20%)',  \n",
    "    grid_gap='20px 20px'\n",
    "))\n",
    "grid_sliders = GridBox(sliders, layout=Layout(\n",
    "    width='60%',\n",
    "    grid_template_columns='repeat(4, 20%)',  \n",
    "    grid_gap='20px 20px'\n",
    "))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def save_button(b):\n",
    "    \"\"\"button to save the sliders values. Potentially normalize values if the dropdown is selected. \n",
    "\n",
    "    Args:\n",
    "        b (array): automatically done by ipywidgets\n",
    "    \"\"\"\n",
    "    node1_value = int(node1.value)\n",
    "    node2_value = int(node2.value)\n",
    "    axx = axx_slider.value\n",
    "    axy = axy_slider.value\n",
    "    bxx = bxx_slider.value\n",
    "    bxy = bxy_slider.value\n",
    "    ayx = ayx_slider.value\n",
    "    ayy = ayy_slider.value\n",
    "    byx = byx_slider.value\n",
    "    byy = byy_slider.value\n",
    "    cxx = cxx_slider.value\n",
    "    cxy = cxy_slider.value\n",
    "    dxx = dxx_slider.value\n",
    "    dxy = dxy_slider.value\n",
    "    cyx = cyx_slider.value\n",
    "    cyy = cyy_slider.value\n",
    "    dyx = dyx_slider.value\n",
    "    dyy = dyy_slider.value\n",
    "    fxy = fxy_slider.value\n",
    "    fyx = fyx_slider.value\n",
    "    gxy = gxy_slider.value\n",
    "    gyx = gyx_slider.value\n",
    "    xi0 = xi0_slider.value\n",
    "    yi0 = yi0_slider.value\n",
    "    xp0 = xp0_slider.value\n",
    "    yp0 = yp0_slider.value\n",
    "\n",
    "    if (node1_value == node2_value):\n",
    "        statustext.value = \"Error: Select two different people\"\n",
    "        return\n",
    "    G[int(node1_value)][int(node2_value)]['parameters'][0] = np.array([[axx, axy, bxx, bxy], [ayx, ayy, byx, byy], [cxx, cxy, dxx, dxy], [cyx, cyy, dyx, dyy]])\n",
    "    G[int(node1_value)][int(node2_value)]['parameters'][1] = np.array([fxy, fyx, gxy, gyx])\n",
    "    G[int(node1_value)][int(node2_value)]['parameters'][2] = np.array([xi0, yi0, xp0, yp0])\n",
    "\n",
    "    if (normalizecheckbox.value):\n",
    "        normalize_values()\n",
    "\n",
    "    calculatedODE = multiple.calculateODEMatrixVector(G[node1_value][node2_value]['parameters'][0], G[node1_value][node2_value]['parameters'][1], G[node1_value][node2_value]['parameters'][2])\n",
    "    G[node1_value][node2_value][\"data\"] = calculatedODE.y\n",
    "    G[node1_value][node2_value][\"t\"] = calculatedODE.t\n",
    "    if (normalize1checkbox.value and normalizecheckbox.value):\n",
    "        statustext.value = \"Success: Saved parameters, normalized values with sum=0, scaled normalized values to -1 or 1, updated sliders, updated plot\"\n",
    "    elif (normalizecheckbox.value):\n",
    "        statustext.value = \"Success: Saved parameters, normalized values with sum=0, updated sliders, updated plot\"\n",
    "    else:\n",
    "        statustext.value = \"Success: Saved parameters, updated sliders, and updated plot\"\n",
    "    update_slider_values()\n",
    "    update_plot()\n",
    "\n",
    "def randomize_button(b):\n",
    "    \"\"\"Randomizes the parametrs for the selected people. \n",
    "\n",
    "    Args:\n",
    "        b (array): automatically done by ipywidgets\n",
    "    \"\"\"\n",
    "    node1_value = int(node1.value)\n",
    "    node2_value = int(node2.value)\n",
    "    G[node1_value][node2_value]['parameters'][0] = multiple.random_starting_matrix()\n",
    "    G[node1_value][node2_value]['parameters'][1] = multiple.random_starting_vector()\n",
    "    G[node1_value][node2_value]['parameters'][2] = multiple.random_initial_vector()\n",
    "\n",
    "    if (normalizecheckbox.value):\n",
    "        normalize_values()\n",
    "\n",
    "    calculatedODE = multiple.calculateODEMatrixVector(G[node1_value][node2_value]['parameters'][0], G[node1_value][node2_value]['parameters'][1], G[node1_value][node2_value]['parameters'][2])\n",
    "    G[node1_value][node2_value][\"data\"] = calculatedODE.y\n",
    "    G[node1_value][node2_value][\"t\"] = calculatedODE.t\n",
    "    statustext.value = \"Success: Randomized parameters for selected people. Parameters were saved, sliders updated, and plot updated.\"\n",
    "    update_slider_values()\n",
    "    update_plot()\n",
    "\n",
    "def randomize_all_button(b):\n",
    "    \"\"\"randomizes parameters across all network edges\n",
    "\n",
    "    Args:\n",
    "        b (array): automatically done by ipywidgets\n",
    "    \"\"\"\n",
    "    multiple.randomize_parameter_network()\n",
    "    if (normalizecheckbox.value):\n",
    "        normalize_values()\n",
    "    statustext.value = \"Success: Randomized parameters for all people. Parameters were saved, sliders updated, and plot updated.\"\n",
    "    update_slider_values()\n",
    "    update_plot()\n",
    "\n",
    "def update_plot_button(b):\n",
    "    \"\"\"updates the plots without actually saving parameters or ODE results of new parameters to the network. \n",
    "\n",
    "    Args:\n",
    "        b (array): automatically done by ipywidgets\n",
    "    \"\"\"\n",
    "    matrixA = np.array([[axx_slider.value, axy_slider.value, bxx_slider.value, bxy_slider.value], [ayx_slider.value, ayy_slider.value, byx_slider.value, byy_slider.value], [cxx_slider.value, cxy_slider.value, dxx_slider.value, dxy_slider.value], [cyx_slider.value, cyy_slider.value, dyx_slider.value, dyy_slider.value]])\n",
    "    vectorB = np.array([fxy_slider.value, fyx_slider.value, gxy_slider.value, gyx_slider.value])\n",
    "    vectorC = np.array([xi0_slider.value, yi0_slider.value, xp0_slider.value, yp0_slider.value])\n",
    "\n",
    "    calculatedODE = multiple.calculateODEMatrixVector(matrixA, vectorB, vectorC)\n",
    "    y_data  = calculatedODE.y\n",
    "    t_data = calculatedODE.t\n",
    "\n",
    "    update_plot(t_data=t_data, y_data=y_data)\n",
    "\n",
    "# button actions\n",
    "savebutton.on_click(save_button)\n",
    "randomizebutton.on_click(randomize_button)\n",
    "updateplotbutton.on_click(update_plot_button)\n",
    "downloadGbutton.on_click(save_graph)\n",
    "# uploadGfile.observe(upload_graph, names='value')\n",
    "uploadGfilebutton.on_click(upload_graph)\n",
    "\n",
    "# initial display of widgets\n",
    "display(widgets.VBox([grid_dropdown, grid_buttons, grid_checkboxes, grid_sliders])) # grid_select\n",
    "display(out)\n",
    "\n",
    "'''\n",
    "how to effectively use this plot. \n",
    "1. select two people from the dropdown menu. By default, the first two people are selected (perosn 0 and 1). \n",
    "2. adjust the sliders to change the parameters of the differential equation. You can then either save the parameters to the graph object by clicking the \"Save Parameters\" button, or you can check the results of the parameters by clicking the \"Update Plot (not saved)\" button. As the name suggests, the \"Update Plot (not saved)\" button does not save the parameters to the graph object. \n",
    "2a. If you click the \"Save Parameters\" button while the \"Normalize values\" button, the parameters aross the whole graph will be normalized with the condition that the sum of the normalized values is (near) 0. If you click the \"Save Parameters\" button while the \"Normalize values\" and \"Scale values to sum to 1\" buttons are checked, the parameters across the whole graph will be normalized with the condition that the sum of the normalized values is 0. and then scaled such that the maximum of the absolute value of the normalized values is 1 or -1. the normalizatoin is done across all vlaue sof the same type. So all axx values are normalized together, all axy values are normalized together, etc.\n",
    "3. You can randomize parameter values by clicking the \"Randomized These Parameters\" button.\n",
    "You can randomize the whole network by clicking on the \"Randomize All Parameters\" button.\n",
    "4. You can save the graph object to a file (named as \"love_multiple.pickle\") by clicking the \"Save Graph To File\" button.\n",
    "5. You can upload a graph object from a file by typing the file path in the text box (either a relative path from where this file is being run from, or by the full path if the file is stored elsewhere). Then click on the \"Submit file path\" button. The data is then loaded into the graph object.\n",
    "6. Below this graph is another graph where it shows you the output from the main graph but for all of the edges in the network. It is not interactive, but it is a good way to see how the parameters affect the output of the differential equation, to see how the parameters are changing over time, and look for potentially interesting patterns/combinaitons in parameters. The parameters are listed to the right of the graph.\n",
    "6a. If you want to update the list of all plots below the interactive plot you will have to re-run the cell. It should take 10-20 seconds to re-run the cell, hence why its not done automatically. \n",
    "7. The plot color has a time component. The color of the plot is determined by the time value. The time scale is the same for all the plots. \n",
    "8. The 3D plots have the dots projected onto the 2D plane via a + mark colored in red, green, or blue depending on the plane/axis of projection. This is to help aid in visualizing the 3D plot. \n",
    "9. There is an inactive textbox where you can see the console output. This is useful for seeing if there are any errors, and for confirmation that an event happened (such as saving a file, uploading a file, etc.)\n",
    "10. The randomize parameter does not work on the xi0, yi0, xp0, yp0 parameters. This is because these parameters are not used in the differential equation. They are used to determine the initial conditions of the differential equation. Likewise for the normalize parameters.\n",
    "'''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# to spit out all of the grpahs in the network\n",
    "number_edges = len(G.edges)\n",
    "fig = plt.figure(figsize=(25, 300/45 * number_edges)) # 300/45 is a \"good\" height for each row of plots, then sclae it by the number of edges in the network (the number of rows to be shown)\n",
    "counter = 1\n",
    "for i in range(number_of_nodes):\n",
    "    for j in range(i+1, number_of_nodes):\n",
    "        xa = G[i][j][\"data\"]\n",
    "        t = G[i][j][\"t\"]\n",
    "        ax = fig.add_subplot(number_edges, 5, counter)\n",
    "        ax.scatter(xa[0], xa[1], s=50, cmap='viridis', alpha=0.5, c=t)\n",
    "        ax.scatter(xa[2], xa[3], s=50, cmap='viridis', alpha=0.5, c=t)\n",
    "        ax.plot(xa[0],xa[1], c=\"black\", label=\"Intimacy\", alpha=0.5)\n",
    "        ax.plot(xa[2],xa[3], c=\"red\", label=\"Passion\", alpha=0.5)\n",
    "        ax.set_xlabel(\"Person \" + str(i), fontsize=10)\n",
    "        ax.set_ylabel(\"Person \" + str(j), fontsize=10)\n",
    "        ax.grid(True)\n",
    "        ax.legend()\n",
    "        counter += 1\n",
    "\n",
    "        # second subplot\n",
    "        ax = fig.add_subplot(number_edges, 5, counter)\n",
    "        ax.scatter(xa[0], xa[2], s=50, cmap='viridis', alpha=0.5, c=t)\n",
    "        ax.scatter(xa[1], xa[3], s=50, cmap='viridis', alpha=0.5, c=t)\n",
    "        ax.plot(xa[0],xa[2], c=\"black\", label=\"Person \" + str(j), alpha=0.5)\n",
    "        ax.plot(xa[1],xa[3], c=\"red\", label=\"Person \" + str(i), alpha=0.5)\n",
    "        ax.set_xlabel('Intimacy', fontsize=10)\n",
    "        ax.set_ylabel('Passion', fontsize=10)\n",
    "        ax.grid(True)\n",
    "        ax.legend()\n",
    "        counter += 1\n",
    "\n",
    "        # third subplot\n",
    "        ax = fig.add_subplot(number_edges, 5, counter, projection='3d')\n",
    "        ax.scatter(xa[0], xa[1], xa[2], c=t, s=20, cmap='viridis', alpha=0.5)\n",
    "        x_lim = ax.get_xlim()\n",
    "        y_lim = ax.get_ylim()\n",
    "        z_lim = ax.get_zlim()\n",
    "        ax.plot(xa[0], xa[2], 'r+', zdir='y', zs=y_lim[1], alpha=0.5)\n",
    "        ax.plot(xa[1], xa[2], 'g+', zdir='x', zs=x_lim[0], alpha=0.5)\n",
    "        ax.plot(xa[0], xa[1], 'b+', zdir='z', zs=z_lim[0], alpha=0.5)\n",
    "        ax.set_xlabel('Intimacy of Person ' + str(i), fontsize=10)\n",
    "        ax.set_ylabel('Intimacy of Person ' + str(j), fontsize=10)\n",
    "        ax.set_zlabel('Passion of Person ' + str(i), fontsize=10)\n",
    "        ax.grid(True)\n",
    "        counter += 1\n",
    "\n",
    "        # fourth subplot\n",
    "        ax = fig.add_subplot(number_edges, 5, counter, projection='3d')\n",
    "        ax.scatter(xa[0], xa[1], xa[3], c=t, cmap='viridis', alpha=0.5, s=20)\n",
    "        x_lim = ax.get_xlim()\n",
    "        y_lim = ax.get_ylim()\n",
    "        z_lim = ax.get_zlim()\n",
    "        ax.plot(xa[0], xa[3], 'r+', zdir='y', zs=y_lim[1], alpha=0.5)\n",
    "        ax.plot(xa[1], xa[3], 'g+', zdir='x', zs=x_lim[0], alpha=0.5)\n",
    "        ax.plot(xa[0], xa[1], 'b+', zdir='z', zs=z_lim[0], alpha=0.5)\n",
    "        ax.set_xlabel('Intimacy of Person ' + str(i), fontsize=10)\n",
    "        ax.set_ylabel('Intimacy of Person ' + str(j), fontsize=10)\n",
    "        ax.set_zlabel('Passion of Person ' + str(j), fontsize=10)\n",
    "        ax.grid(True)\n",
    "        counter += 1\n",
    "        \n",
    "        # 5th \"plot\", shows the used parameters to offer reproducability\n",
    "        string = \"axx={}, axy={}, bxx={}, bxy={}\\nayx={}, ayy={}, byx={}, byy={}\\ncxx={}, cxy={}, dxx={}, dxy={}\\ncyx={}, cyy={}, dyx={}, dyy={}\\nfxy={}, fyx={}, gxy={}, gyx={}, \\n xi0={}, yi0={}, xp0={}, yp0={}\".format(\n",
    "            \"{:.2f}\".format(G[i][j]['parameters'][0][0][0]), \"{:.2f}\".format(G[i][j]['parameters'][0][0][1]), \"{:.2f}\".format(G[i][j]['parameters'][0][0][2]), \"{:.2f}\".format(G[i][j]['parameters'][0][0][3]), \n",
    "            \"{:.2f}\".format(G[i][j]['parameters'][0][1][0]), \"{:.2f}\".format(G[i][j]['parameters'][0][1][1]), \"{:.2f}\".format(G[i][j]['parameters'][0][1][2]), \"{:.2f}\".format(G[i][j]['parameters'][0][1][3]), \n",
    "            \"{:.2f}\".format(G[i][j]['parameters'][0][2][0]), \"{:.2f}\".format(G[i][j]['parameters'][0][2][1]), \"{:.2f}\".format(G[i][j]['parameters'][0][2][2]), \"{:.2f}\".format(G[i][j]['parameters'][0][2][3]), \n",
    "            \"{:.2f}\".format(G[i][j]['parameters'][0][3][0]), \"{:.2f}\".format(G[i][j]['parameters'][0][3][1]), \"{:.2f}\".format(G[i][j]['parameters'][0][3][2]), \"{:.2f}\".format(G[i][j]['parameters'][0][3][3]), \n",
    "            \"{:.2f}\".format(G[i][j]['parameters'][1][0]), \"{:.2f}\".format(G[i][j]['parameters'][1][1]), \"{:.2f}\".format(G[i][j]['parameters'][1][2]), \"{:.2f}\".format(G[i][j]['parameters'][1][3]), \n",
    "            \"{:.2f}\".format(G[i][j]['parameters'][2][0]), \"{:.2f}\".format(G[i][j]['parameters'][2][1]), \"{:.2f}\".format(G[i][j]['parameters'][2][2]), \"{:.2f}\".format(G[i][j]['parameters'][2][3])\n",
    "            )\n",
    "\n",
    "        ax = fig.add_subplot(number_edges, 5, counter)\n",
    "        ax.text(0.5, 0.5, string, horizontalalignment='center', verticalalignment='center', transform=ax.transAxes)\n",
    "        ax.axis('off')\n",
    "        counter += 1\n",
    "show()"
   ]
  }
 ],
 "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": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}