visual.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "import plotly\n",
    "import plotly.graph_objects as go\n",
    "from plotly.subplots import make_subplots\n",
    "\n",
    "from model import Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def testData():\n",
    "    theta = np.random.random(25) * np.pi * 1.2\n",
    "    r = np.random.random(25) * 0.3 + 0.6\n",
    "\n",
    "    x1 = np.concatenate(((r * np.cos(theta))[:,None], (r * np.sin(theta))[:,None]), axis=1)\n",
    "    y1 = np.ones(25)\n",
    "    \n",
    "    theta = np.random.random(25) * np.pi * 1.2 + 1.2 * np.pi\n",
    "    r = np.random.random(25) * 0.2 + 0.05\n",
    "\n",
    "    x0 = np.concatenate(((r * np.cos(theta))[:,None], (r * np.sin(theta))[:,None]), axis=1)\n",
    "    y0 = np.zeros(25)\n",
    "    \n",
    "    x = np.concatenate((x0, x1), axis=0)\n",
    "    y = np.concatenate((y0, y1), axis=0)\n",
    "    \n",
    "    return x, y[:,None].astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: Train and save data for visualization\n",
    "\n",
    "np.random.seed(35) # \n",
    "\n",
    "x, y = testData()\n",
    "model = Model()\n",
    "\n",
    "data = {'x':[], 'dl':[], 'grid':[], 'acc':[]}\n",
    "\n",
    "xs, dls, grids = model.spaces(x)\n",
    "acc = model.accuracy(x, y)\n",
    "\n",
    "data['x'].append(xs)\n",
    "data['dl'].append(dls)\n",
    "data['grid'].append(grids)\n",
    "data['acc'].append(acc)\n",
    "\n",
    "for t in range(120): \n",
    "    \n",
    "    for i in range(1): \n",
    "        \n",
    "        loss = model.opt(x, y, 0.1)\n",
    "        \n",
    "        acc = model.accuracy(x, y)\n",
    "\n",
    "    # print(f\"{loss = :<10.8} {acc = :.4%}\")\n",
    "\n",
    "    xs, dls, grids = model.spaces(x)\n",
    "    \n",
    "    data['x'].append(xs)\n",
    "    data['dl'].append(dls)\n",
    "    data['grid'].append(grids)\n",
    "    data['acc'].append(acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# make subplots\n",
    "fig = make_subplots(\n",
    "    rows=1,\n",
    "    cols=3,\n",
    "    subplot_titles=(\n",
    "        \"Orignal Space\",\n",
    "        \"Latent Space 1\",\n",
    "        \"Latent Space 2\",\n",
    "        \"Latent Space 3\",\n",
    "    ),\n",
    "    horizontal_spacing=0.05,\n",
    "    vertical_spacing=0.01,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# the initial plot\n",
    "t = 1\n",
    "\n",
    "xs = data[\"x\"][t]\n",
    "dls = data[\"dl\"][t]\n",
    "grids = data[\"grid\"][t]\n",
    "\n",
    "for i in range(3):\n",
    "\n",
    "    for j in range(len(grids[i])):\n",
    "        fig.add_trace(\n",
    "            go.Scatter(\n",
    "                x=grids[i][j][:, 0],\n",
    "                y=grids[i][j][:, 1],\n",
    "                mode=\"lines\",\n",
    "                marker_color=\"lightgreen\",\n",
    "                line={\"dash\": \"dash\"},\n",
    "                hoverinfo='none',\n",
    "            ),\n",
    "            row=1,\n",
    "            col=i + 1,\n",
    "        )\n",
    "\n",
    "    fig.add_traces(\n",
    "        [\n",
    "            go.Scatter(\n",
    "                x=xs[i][:, 0],\n",
    "                y=xs[i][:, 1],\n",
    "                mode=\"markers\",\n",
    "                marker_color=list(map(lambda x: \"orange\" if x else \"blue\", y)),\n",
    "                hoverinfo='skip',\n",
    "            ),\n",
    "            go.Scatter(x=dls[i][:, 0], y=dls[i][:, 1], mode=\"lines\", line_color=\"red\", hoverinfo='skip'),\n",
    "        ],\n",
    "        rows=1,\n",
    "        cols=i + 1,\n",
    "    )\n",
    "\n",
    "# build each frame\n",
    "frames = []\n",
    "\n",
    "for t in range(len(data[\"x\"])):\n",
    "\n",
    "    xs = data[\"x\"][t]\n",
    "    dls = data[\"dl\"][t]\n",
    "    grids = data[\"grid\"][t]\n",
    "\n",
    "    frame = []\n",
    "\n",
    "    for i in range(3):\n",
    "\n",
    "        for j in range(len(grids[i])):\n",
    "            frame.append(\n",
    "                go.Scatter(\n",
    "                    x=grids[i][j][:, 0],\n",
    "                    y=grids[i][j][:, 1],\n",
    "                    mode=\"lines\",\n",
    "                    marker_color=\"lightgreen\",\n",
    "                    line={\"dash\": \"dash\"}, \n",
    "                    hoverinfo='none',\n",
    "                )\n",
    "            )\n",
    "\n",
    "        frame.extend(\n",
    "            [\n",
    "                go.Scatter(\n",
    "                    x=xs[i][:, 0],\n",
    "                    y=xs[i][:, 1],\n",
    "                    mode=\"markers\",\n",
    "                    marker_color=list(map(lambda x: \"orange\" if x else \"blue\", y)),\n",
    "                    hoverinfo='skip', \n",
    "                ),\n",
    "                go.Scatter(\n",
    "                    x=dls[i][:, 0], y=dls[i][:, 1], mode=\"lines\", line_color=\"red\", hoverinfo='skip'\n",
    "                ),\n",
    "            ]\n",
    "        )\n",
    "\n",
    "    frames.append(go.Frame(data=frame, traces=list(range(36)), name=str(t))) # Name used to link slider\n",
    "\n",
    "fig.frames = frames\n",
    "\n",
    "# build button\n",
    "button_dict = {\n",
    "    \"buttons\": [\n",
    "        {\n",
    "            \"args\": [\n",
    "                None,\n",
    "                {\n",
    "                    \"frame\": {\"duration\": 50, \"redraw\": False},\n",
    "                    \"fromcurrent\": True,\n",
    "                    \"transition\": {\"duration\": 10},\n",
    "                },\n",
    "            ],\n",
    "            \"label\": \"Run\",\n",
    "            \"method\": \"animate\",\n",
    "        },\n",
    "        {\n",
    "            \"args\": [\n",
    "                [None],\n",
    "                {\n",
    "                    \"frame\": {\"duration\": 0, \"redraw\": False},\n",
    "                    \"mode\": \"immediate\",\n",
    "                    \"transition\": {\"duration\": 0},\n",
    "                },\n",
    "            ],\n",
    "            \"label\": \"Pause\",\n",
    "            \"method\": \"animate\",\n",
    "        },\n",
    "    ],\n",
    "    \"direction\": \"left\",\n",
    "    \"pad\": {\"r\": 10, \"t\": 87},\n",
    "    \"showactive\": False,\n",
    "    \"type\": \"buttons\",\n",
    "    \"x\": 0.1,\n",
    "    \"xanchor\": \"right\",\n",
    "    \"y\": 0.05,\n",
    "    \"yanchor\": \"top\",\n",
    "}\n",
    "\n",
    "fig.update_layout(updatemenus=[button_dict])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sliders_dict = {\n",
    "    \"active\": 0,\n",
    "    \"yanchor\": \"top\",\n",
    "    \"xanchor\": \"left\",\n",
    "    \"currentvalue\": {\n",
    "        \"font\": {\"size\": 20},\n",
    "        \"prefix\": \"Step:\",\n",
    "        \"visible\": True,\n",
    "        \"xanchor\": \"right\"\n",
    "    },\n",
    "    \"transition\": {\"duration\": 10},\n",
    "    \"pad\": {\"b\": 10, \"t\": 50},\n",
    "    \"len\": 0.9,\n",
    "    \"x\": 0.1,\n",
    "    \"y\": 0.05,\n",
    "    \"steps\": []\n",
    "}\n",
    "\n",
    "for t in range(len(data[\"x\"])):\n",
    "    slider_step = {\"args\": [\n",
    "        [t],\n",
    "        {\"frame\": {\"duration\": 50, \"redraw\": False},\n",
    "         \"mode\": \"immediate\",\n",
    "         \"transition\": {\"duration\": 10}}\n",
    "    ],\n",
    "        \"label\": t,\n",
    "        \"method\": \"animate\"}\n",
    "    sliders_dict[\"steps\"].append(slider_step)\n",
    "\n",
    "fig.update_layout(updatemenus=[button_dict], sliders=[sliders_dict])\n",
    "\n",
    "fig.update_layout(width=1200, height=550, showlegend=False, title=\"Visualize (Latent) Spaces during Training of Neural Network\",)\n",
    "\n",
    "fig.update_xaxes(range=[-1.05, 1.05])\n",
    "fig.update_yaxes(range=[-1.05, 1.05])\n",
    "\n",
    "plotly.offline.plot(fig, filename='visual.html');\n",
    "\n",
    "# fig.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Study",
   "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": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}