notebooks/experiments/system_model_v3/notebook-controller-run-analyze-grouped.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Controller Run and Analyze w/ Grouped Output\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup and Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set project root folder, to enable importing project files from subdirectories\n",
    "from pathlib import Path\n",
    "import os\n",
    "\n",
    "path = Path().resolve()\n",
    "root_path = str(path).split('notebooks')[0]\n",
    "os.chdir(root_path)\n",
    "\n",
    "# Import all shared dependencies and setup\n",
    "from shared import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime\n",
    "import time\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "import plotly.express as px\n",
    "import plotly.graph_objects as go\n",
    "from plotly.subplots import make_subplots\n",
    "import plotly.io as pio\n",
    "pio.renderers.default = \"png\"\n",
    "png_renderer = pio.renderers[\"png\"]\n",
    "png_renderer.width = 2000\n",
    "png_renderer.height = 1200\n",
    "\n",
    "\n",
    "from models.system_model_v3.model.params.init import params\n",
    "from experiments.system_model_v3.run import run_experiment\n",
    "from experiments.system_model_v3.post_process import post_process_results\n",
    "from models.constants import RAY "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Simulation Configuration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Number of timesteps(hours) to run\n",
    "# Max timesteps is 24*30*12 = 1 year\n",
    "SIMULATION_TIMESTEPS = 24 * 30 * 6\n",
    "\n",
    "# Number of runs. Each run uses a different simulated ETH dataset\n",
    "MONTE_CARLO_RUNS = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set param values. These will override defaults\n",
    "# For details on params, see REF\n",
    "params_override = { \n",
    "    'debug': [False],\n",
    "    'eth_trend': [1],# 0=no trend; >0=uptrend; <0=downtrend \n",
    "    'liquidity_demand_enabled': [False],\n",
    "    'liquidity_demand_shock': [False],\n",
    "    'liquidation_buffer': [2],\n",
    "    'max_redemption_rate': [50], # used by SAFE owners\n",
    "    'min_redemption_rate': [-50], # used by SAFE owners\n",
    "    'kp': [5e-8],\n",
    "    'ki': [5e-14],\n",
    "    #'alpha': [1.0 * RAY],\n",
    "    'alpha': [0.9999 * RAY],\n",
    "    'rate_trader_mean_pct': [3],\n",
    "    'rate_trader_min_pct': [0],\n",
    "    'rate_trader_std_pct': [2 * (3-0)],\n",
    "    'rate_trader_mean_days': [0],\n",
    "    'rate_trader_min_days': [0],\n",
    "    'rate_trader_std_days': [2 * (0-0)],\n",
    "    'eth_leverager_target_min_liquidity_ratio': [2.9],\n",
    "    'eth_leverager_target_max_liquidity_ratio': [2.9]\n",
    "}\n",
    "params.update(params_override)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run Simulation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "start = time.time()\n",
    "df_raw = run_experiment(timesteps=SIMULATION_TIMESTEPS,\n",
    "               runs=MONTE_CARLO_RUNS, params=params);\n",
    "df = post_process_results(df_raw, params)\n",
    "print(f\"Run experiment and post-process took {time.time() - start} secs\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Optionally, trim results by timestep\n",
    "df_trim = df[df['timestep'] >= 20][df['timestep'] <= 24*30*12]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def group_plot(df, run):\n",
    "    \"\"\"\n",
    "    Show many plots from a single simulation run\n",
    "    \"\"\"\n",
    "    \n",
    "    fig = make_subplots(rows=5, cols=3, start_cell=\"top-left\", vertical_spacing=0.06, \n",
    "                        horizontal_spacing=0.06,\n",
    "                        subplot_titles=[\"ETH/USD\",\"Kp rate(blue), Ki rate(red)\", \"Redemption Rate\", \n",
    "                                        \"RAI/USD\", \"Error\",  \"Rate Trader balances, RAI(purple), USD(green)\",\n",
    "                                        \"Rate APY\", \"Error Integral\", \"ETH Leverager Collateral\",\n",
    "                                        \"Total Debt\", \"Total Collateral\", \"Total C-Ratio\",\n",
    "                                        \"RAI/ETH\"])\n",
    "   \n",
    "    eth_usd = px.line(\n",
    "            df.query(f'run == {run}'),\n",
    "            title=f\"ETH/USD\",\n",
    "            x=\"timestamp\",\n",
    "            color=\"run\",\n",
    "            color_discrete_sequence=[\"blue\"],\n",
    "            labels={\"run\": \"\"},\n",
    "            y=[\"eth_price\"]\n",
    "        )\n",
    "    eth_usd.data[0].name = \"ETH/USD\"\n",
    "    eth_usd.update_layout(title_x=0.5)\n",
    "    \n",
    "    rai_eth = px.line(\n",
    "            df.query(f'run == {run}'),\n",
    "            title=f\"RAI/ETH\",\n",
    "            x=\"timestamp\",\n",
    "            color=\"run\",\n",
    "            color_discrete_sequence=[\"green\"],\n",
    "            labels={\"run\": \"\"},\n",
    "            y=[\"rai_eth\"]\n",
    "        )\n",
    "    rai_eth.data[0].name = \"RAI/ETH\"\n",
    "    rai_eth.update_layout(title_x=0.5)\n",
    "\n",
    "    rai_usd = px.line(\n",
    "            df.query(f'run == {run}'),\n",
    "            title=f\"RAI/USD\",\n",
    "            x=\"timestamp\",\n",
    "            color_discrete_sequence=[\"purple\", \"black\"],\n",
    "            y=[\"market_price_twap\", \"target_price\"]\n",
    "        )\n",
    "    rai_usd.data[0].name = \"RAI/USD TWAP\"\n",
    "    rai_usd.data[1].name = \"Redemption Price\"\n",
    "\n",
    "    rai_apy = px.line(\n",
    "            df.query(f'run == {run}'),\n",
    "            title=f\"Redemption Rate APY\",\n",
    "            x=\"timestamp\",\n",
    "            color_discrete_sequence=[\"blue\"],\n",
    "            y=[\"apy\"]\n",
    "        )\n",
    "    rai_apy.data[0].name = \"Redemption Rate(APY)\"\n",
    "\n",
    "    leverager_collateral = px.line(\n",
    "            df.query(f'run == {run}'),#.query(f'khow == 7e-14'),\n",
    "            title=f\"ETH Leverager Collateral\",\n",
    "            x=\"timestamp\",\n",
    "            color_discrete_sequence=['red'],\n",
    "            y=['eth_leverager_collateral']\n",
    "        )\n",
    "    leverager_collateral.data[0].name = \"ETH Leverager Collateral\"\n",
    "    leverager_collateral.update_layout(title_x=0.5)\n",
    "\n",
    "    leverager_collateral_diff = px.line(\n",
    "            df.query(f'run == {run}'),#.query(f'khow == 7e-14'),\n",
    "            title=f\"ETH Leverager Collateral Diff\",\n",
    "            x=\"timestamp\",\n",
    "            color_discrete_sequence=['red'],\n",
    "            y=['eth_leverager_collateral_diff']\n",
    "        )\n",
    "    leverager_collateral_diff.data[0].name = \"ETH Leverager Collateral Diff\"\n",
    "    leverager_collateral_diff.update_layout(title_x=0.5)\n",
    "\n",
    "    rate_trader = px.line(\n",
    "            df.query(f'run == {run}'),#.query(f'khow == 7e-14'),\n",
    "            title=f\"Rate Trader balances\",\n",
    "            x=\"timestamp\",\n",
    "            color_discrete_sequence=['green', 'purple'],\n",
    "            y=['rate_trader_base', 'rate_trader_total_rai_base']\n",
    "        )\n",
    "    rate_trader.data[0].name = \"Rate Trader Base only\"\n",
    "    rate_trader.data[1].name = \"Rate Trader RAI(Base)\"\n",
    "    rate_trader.update_layout(title_x=0.5)\n",
    "\n",
    "    rate_comps = px.line(\n",
    "            df.query(f'run == {run}'),\n",
    "            title=f\"Errors\",\n",
    "            x=\"timestamp\",\n",
    "            color_discrete_sequence=[\"blue\", \"red\"],\n",
    "            y=[\"kp_rate\", \"ki_rate\"]\n",
    "        )\n",
    "    rate_comps.data[0].name = \"Kp rate\"\n",
    "    rate_comps.data[1].name = \"Ki rate\"\n",
    "    \n",
    "    ki_rate = px.line(\n",
    "            df.query(f'run == {run}'),\n",
    "            title=f\"Errors\",\n",
    "            x=\"timestamp\",\n",
    "            color_discrete_sequence=[\"red\"],\n",
    "            y=[\"ki_rate\"]\n",
    "        )\n",
    "    ki_rate.data[0].name = \"Ki rate\"\n",
    "\n",
    "    error = px.line(\n",
    "            df.query(f'run == {run}'),\n",
    "            title=f\"Errors\",\n",
    "            x=\"timestamp\",\n",
    "            color_discrete_sequence=[\"blue\"],\n",
    "            y=[\"error_star\"]\n",
    "        )\n",
    "    error.data[0].name = \"Error\"\n",
    "\n",
    "    error_integral = px.line(\n",
    "            df.query(f'run == {run}'),\n",
    "            title=f\"Errors\",\n",
    "            x=\"timestamp\",\n",
    "            color_discrete_sequence=[\"red\"],\n",
    "            y=[\"error_star_integral\"]\n",
    "        )\n",
    "    error_integral.data[0].name = \"Error Integral\"\n",
    "    \n",
    "    \n",
    "    target_rate = px.line(\n",
    "            df.query(f'run == {run}'),\n",
    "            title=f\"RR\",\n",
    "            x=\"timestamp\",\n",
    "            color_discrete_sequence=[\"black\"],\n",
    "            y=[\"target_rate\"]\n",
    "    )\n",
    "    target_rate.data[0].name = \"Redemption Rate\"\n",
    "\n",
    "    eth_leverager_cratio = px.line(\n",
    "            df.query(f'run == {run}'),\n",
    "            title=f\"ETH Leverager C-ratio\",\n",
    "            x=\"timestamp\",\n",
    "            color_discrete_sequence=['blue'],\n",
    "            y=['eth_leverager_cratio']\n",
    "        )\n",
    "    eth_leverager_cratio.data[0].name = \"ETH Leverager C-ratio\"\n",
    "    eth_leverager_cratio.update_layout(title_x=0.5)\n",
    "    \n",
    "    \n",
    "    total_debt = px.line(\n",
    "        df_trim.query(f'run == {run}'), \n",
    "        title=\"Total Debt\",\n",
    "        x=\"timestamp\", \n",
    "        y=[\"principal_debt\"], \n",
    "        color_discrete_sequence=['purple']\n",
    "    )\n",
    "    total_collateral = px.line(\n",
    "        df_trim.query(f'run == {run}'), \n",
    "        title=\"Total collateral\",\n",
    "        x=\"timestamp\", \n",
    "        y=[\"eth_collateral\"], \n",
    "        color_discrete_sequence=['red']\n",
    "    )\n",
    "    \n",
    "    total_cratio = px.line(\n",
    "        df_trim.query(f'run == {run}'), \n",
    "        title=\"Total C-ratio\",\n",
    "        x=\"timestamp\", \n",
    "        y=[\"collateralization_ratio\"], \n",
    "        color_discrete_sequence=['orange']\n",
    "    )\n",
    "    total_cratio.update_layout(font={'size': 40})\n",
    "    \n",
    "    fig.update_layout(title_x=0.5)\n",
    "\n",
    "    # First column\n",
    "    fig.add_trace(eth_usd.data[0], row=1, col=1)\n",
    "    fig.add_trace(rai_usd.data[0], row=2, col=1)\n",
    "    fig.add_trace(rai_usd.data[1], row=2, col=1)\n",
    "    fig.add_trace(rai_apy.data[0], row=3, col=1)\n",
    "    fig.add_trace(total_debt.data[0], row=4, col=1)\n",
    "    fig.add_trace(rai_eth.data[0], row=5, col=1)\n",
    "\n",
    "    # Second column\n",
    "    fig.add_trace(rate_comps.data[0], row=1, col=2)\n",
    "    fig.add_trace(rate_comps.data[1], row=1, col=2)\n",
    "    fig.add_trace(error.data[0], row=2, col=2)\n",
    "    fig.add_trace(error_integral.data[0], row=3, col=2)\n",
    "    fig.add_trace(total_collateral.data[0], row=4, col=2)\n",
    "    \n",
    "    # Third Column\n",
    "    fig.add_trace(target_rate.data[0], row=1, col=3)\n",
    "    fig.add_trace(rate_trader.data[0], row=2, col=3)\n",
    "    fig.add_trace(rate_trader.data[1], row=2, col=3)\n",
    "    fig.add_trace(leverager_collateral.data[0], row=3, col=3)\n",
    "    fig.add_trace(total_cratio.data[0], row=4, col=3)\n",
    "\n",
    "    fig.update_traces(line=dict(width=2))\n",
    "    fig.update_layout(showlegend=False)\n",
    "\n",
    "    # only show tick labels at bottom\n",
    "    fig.update_xaxes(showticklabels=False) # hide all the xticks\n",
    "    fig.update_xaxes(showticklabels=True, row=5, col=1)\n",
    "    fig.update_xaxes(showticklabels=True, row=4, col=2)\n",
    "    fig.update_xaxes(showticklabels=True, row=4, col=3)\n",
    "\n",
    "    fig.update_annotations(font_size=24)\n",
    "    fig.update_xaxes(\n",
    "       tickfont = dict(size=18))\n",
    "\n",
    "    fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for run in range(1, MONTE_CARLO_RUNS + 1):\n",
    "    print(f\"{run=}\")\n",
    "    group_plot(df_trim, run)"
   ]
  }
 ],
 "metadata": {
  "jupytext": {
   "cell_metadata_filter": "-all",
   "encoding": "# -*- coding: utf-8 -*-",
   "formats": "ipynb,py:percent"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}