Uniswap_Model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Simple Uniswap Demo\n",
    "\n",
    "## Introduction to Uniswap\n",
    ">Uniswap is an automated market maker for exchanging ERC20 tokens. Anyone can become a liquidity provider, and invest in the liquidity pool of an ERC20 token. This allows other users to trade that token for other tokens at an exchange rate based on their relative availibility. When a token trade is executed, a small fee is paid to the liquidity providers that enabled the transaction.\n",
    "https://uniswap.io/\n",
    "\n",
    "There are basically eight ways (_mechanisms_) in which an agent can interact with an instance of Uniswap (https://github.com/Uniswap/uniswap-v1/blob/master/contracts/uniswap_exchange.vy):\n",
    "1. `addLiquidity`: deposit ETH and tokens in the liquidity pool; mint UNI tokens - \"shares\" of that Uniswap instance - in exchange;\n",
    "1. `removeLiquidity`: burn UNI tokens; withdraw a proportional amount of ETH and tokens in exchange;\n",
    "1. `ethToTokenInput`: user specifies an exact amount of ETH they send; receives corresponding amount of tokens\n",
    "1. `ethToTokenOutput`: user sends some ETH and specifies an exact amount of tokens they want to buy; Uniswap refunds ETH sent in excess\n",
    "1. `tokenToEthInput`: user specifies an exact amount of tokens they send; receives corresponding amount of ETH\n",
    "1. `tokenToEthOutput`: user specifies an exact amount of ETH they want to buy; Uniswap takes the corresponding amount of tokens from the user account\n",
    "1. `tokenToTokenInput`: user specifies an exact amount of \"token A\" they send; receives corresponding amount of \"token B\" (effectively a `tokenToEthInput` in instance A combined with a `ethToTokenInput` in instance B)\n",
    "1. `tokenToTokenOutput`: user specifies an exact amount of \"token B˜ they want to buy; Uniswap takes the corresponding amount of \"token A\" from the user account (effectively a `ethToTokenOutput` in instance B combined with a `tokenToEthOutput` in instance A)\n",
    "\n",
    "## Limitations and simplifications of this model\n",
    "* When triggering mechanisms, Uniswap users may include conditions they would like to apply to the transaction. For example, they could define a minimum amount of UNI tokens they wish to receive as a result of adding liquidity to the pool; or the maximum number of tokens they are willing to pay for the amount of ETH requested in a `tokenToEthOutput` operation; or a deadline after which the transaction should not be processed. **This simplified model does not account for this kind conditioning.**\n",
    "\n",
    "* The `ethToTokenOutput`, `tokenToEthOutput` mechanisms are not implemented. **All swaps are treated as of the \"input defined\" type.**\n",
    "\n",
    "* The `tokenToToken` mechanism are not implemented, as they are merely a `tokenToEth` in the context of a single Uniswap instance.\n",
    "\n",
    "* **User _behavior_ has not been modeled**. User _actions_ are derived from the history of events of the Uniswap instance being analyzed.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to better mimic the behavior of Uniswap, we need to pay special attention to two things:\n",
    "1. make explicit that we want numbers to be treated as integers by the Python interpreter - there are probably more casts to `int` than needed throughout the code;\n",
    "1. Uniswap was coded in Vyper, where the `/` operator applied to integer variables returns the dividend of the division. In order to achieve the same in Python, we need to use _\"floor division\"_ (`//`) instead of _\"true division\"_ (`/`)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000000000000000000 / 5000000000000000001 = 2.0\n",
      "int(10000000000000000000 / 5000000000000000001) = 2\n",
      "10000000000000000000 // 5000000000000000001 = 1\n"
     ]
    }
   ],
   "source": [
    "expr = '10000000000000000000 / 5000000000000000001'\n",
    "print(f'{expr} = {eval(expr)}')\n",
    "expr = 'int(10000000000000000000 / 5000000000000000001)'\n",
    "print(f'{expr} = {eval(expr)}')\n",
    "expr = '10000000000000000000 // 5000000000000000001'\n",
    "print(f'{expr} = {eval(expr)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model File structure\n",
    "* Uniswap_Model.ipynb\n",
    "* model\n",
    "\n",
    "In the model folder there exist 4 files and a parts folder, the [config.py](model/config.py), [partial_state_update_block.py](model/partial_state_update_block.py), [run.py](model/run.py), and [state_variables.py](state_variables.py). The [config.py](model/config.py) contains the simulation configurations, aggregating the partial states, and the state variables. The [partial_state_update_block.py](model/partial_state_update_block.py) contains the partial state update blocks and how they update the state variables. [state_variables.py](state_variables.py) defines the state variables and [run.py](model/run.py) actually runs the simulation. \n",
    "\n",
    "The components of the model live within the parts subfolder as:\n",
    "\n",
    "* [system.py](model/parts/system.py)\n",
    "* [utils.py](model/parts/utils.py)\n",
    "* uniswap_events.pickle* \n",
    "\n",
    "The initial parameters and hyperparameters of the system are defined in [sys_params.py](model/sys_params.py) and helper functions, plots, etc. are in the [utils.py](model/parts/utils.py).\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model dataset*\n",
    "To run our simulation, we use real uniswap event data that was was generated using [Kaggle/BigQuery](https://www.kaggle.com/markusbkoch/uniswap-ethdai-exchange/). Please download [here](https://www.kaggleusercontent.com/kf/21051736/eyJhbGciOiJkaXIiLCJlbmMiOiJBMTI4Q0JDLUhTMjU2In0..HdZwILo7rBzA0baBSYIogA.5sWqgw2JU7MhBWp6MFRE2kpUJPQyAsOmljPX3jRGeEJMY5bYdZNjeW6XsuOkVDe8v4wcL5Efa3EpkrrC7bu-RLKqEEHazP2L5IQHGoYhschgBumjaoR2VPGM4xL40abyFTesKlVHGef2EIIlwpvHX0DcNrMQI-MpyVEjMSHx_oY.kBNFGv1rap1qLe-tfqQA6A/uniswap_events.pickle). and put in the model/parts folder (the same directory that has the *system.py* and *utils.py* files).\n",
    "\n",
    "To learn how to download and analyze Uniswap data, visit [this notebook](data/ETHDAI_EDA_and_ETL.ipynb)\n",
    "\n",
    "## Run Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting cadCAD\n",
      "  Downloading cadCAD-0.4.23-py3-none-any.whl (76 kB)\n",
      "\u001b[K     |████████████████████████████████| 76 kB 1.2 MB/s eta 0:00:01\n",
      "\u001b[?25hCollecting funcy\n",
      "  Using cached funcy-1.15-py2.py3-none-any.whl (32 kB)\n",
      "Collecting fn\n",
      "  Using cached fn-0.4.3.tar.gz (38 kB)\n",
      "Collecting pathos\n",
      "  Using cached pathos-0.2.7-py2.py3-none-any.whl (81 kB)\n",
      "Requirement already satisfied: pandas in /Users/lana/anaconda3/lib/python3.7/site-packages (from cadCAD) (0.24.2)\n",
      "Collecting dill>=0.3.3\n",
      "  Using cached dill-0.3.3-py2.py3-none-any.whl (81 kB)\n",
      "Collecting multiprocess>=0.70.11\n",
      "  Using cached multiprocess-0.70.11-py3-none-any.whl (98 kB)\n",
      "Collecting pox>=0.2.9\n",
      "  Using cached pox-0.2.9-py2.py3-none-any.whl (30 kB)\n",
      "Collecting ppft>=1.6.6.3\n",
      "  Using cached ppft-1.6.6.3-py3-none-any.whl (65 kB)\n",
      "Requirement already satisfied: numpy>=1.12.0 in /Users/lana/anaconda3/lib/python3.7/site-packages (from pandas->cadCAD) (1.19.0)\n",
      "Requirement already satisfied: python-dateutil>=2.5.0 in /Users/lana/anaconda3/lib/python3.7/site-packages (from pandas->cadCAD) (2.8.1)\n",
      "Requirement already satisfied: pytz>=2011k in /Users/lana/anaconda3/lib/python3.7/site-packages (from pandas->cadCAD) (2018.5)\n",
      "Requirement already satisfied: six>=1.7.3 in /Users/lana/anaconda3/lib/python3.7/site-packages (from ppft>=1.6.6.3->pathos->cadCAD) (1.15.0)\n",
      "Building wheels for collected packages: fn\n",
      "  Building wheel for fn (setup.py) ... \u001b[?25ldone\n",
      "\u001b[?25h  Created wheel for fn: filename=fn-0.4.3-py3-none-any.whl size=28473 sha256=c9a937f6912dadd8ebb173662c7b2715258dc41bdf07bfe2d8bfdbbfa3d0e01e\n",
      "  Stored in directory: /Users/lana/Library/Caches/pip/wheels/5c/a0/65/41f733d04386c57826c2cf9e1927d716e90a46689712c9af6b\n",
      "Successfully built fn\n",
      "\u001b[31mERROR: apache-airflow 1.10.4rc3 has requirement dill<0.3,>=0.2.2, but you'll have dill 0.3.3 which is incompatible.\u001b[0m\n",
      "\u001b[31mERROR: apache-airflow 1.10.4rc3 has requirement flask<2.0,>=1.1.0, but you'll have flask 0.12.5 which is incompatible.\u001b[0m\n",
      "Installing collected packages: funcy, fn, dill, multiprocess, pox, ppft, pathos, cadCAD\n",
      "  Attempting uninstall: dill\n",
      "    Found existing installation: dill 0.2.9\n",
      "    Uninstalling dill-0.2.9:\n",
      "      Successfully uninstalled dill-0.2.9\n",
      "Successfully installed cadCAD-0.4.23 dill-0.3.3 fn-0.4.3 funcy-1.15 multiprocess-0.70.11 pathos-0.2.7 pox-0.2.9 ppft-1.6.6.3\n",
      "\u001b[33mWARNING: You are using pip version 20.1.1; however, version 20.2.4 is available.\n",
      "You should consider upgrading via the '/Users/lana/anaconda3/bin/python -m pip install --upgrade pip' command.\u001b[0m\n"
     ]
    }
   ],
   "source": [
    "!pip install cadCAD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "                  ___________    ____\n",
      "  ________ __ ___/ / ____/   |  / __ \\\n",
      " / ___/ __` / __  / /   / /| | / / / /\n",
      "/ /__/ /_/ / /_/ / /___/ ___ |/ /_/ /\n",
      "\\___/\\__,_/\\__,_/\\____/_/  |_/_____/\n",
      "by cadCAD\n",
      "\n",
      "Execution Mode: local_proc\n",
      "Configuration Count: 1\n",
      "Dimensions of the first simulation: (Timesteps, Params, Runs, Vars) = (147437, 2, 1, 3)\n",
      "Execution Method: local_simulations\n",
      "SimIDs   : [0]\n",
      "SubsetIDs: [0]\n",
      "Ns       : [0]\n",
      "ExpIDs   : [0]\n",
      "Execution Mode: single_threaded\n",
      "Total execution time: 8.48s\n"
     ]
    }
   ],
   "source": [
    "# Dependences\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Experiments\n",
    "from model import run\n",
    "from model.parts.utils import *\n",
    "pd.options.display.float_format = '{:.2f}'.format\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "df = run.run()\n",
    "rdf = run.postprocessing(df)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analysis\n",
    "\n",
    "For information on how to access and analysis data from the Eth/Dai Uniswap smart contract, visit our sister notebook [here](Uniswap_EthDai_Exchange_Data.ipynb).\n",
    "\n",
    "\n",
    "In this section, we will analyze how well our model compares to the actual data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_history = rdf[['DAI_balance', 'ETH_balance', 'UNI_supply']]\n",
    "model_history.columns = ['model_DAI_balance', 'model_ETH_balance', 'model_UNI_supply']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "uniswap_events = pd.read_pickle('model/parts/uniswap_events.pickle')\n",
    "real_history = uniswap_events.iloc[1:][['token_balance','eth_balance','UNI_supply']].reset_index(drop=True)\n",
    "real_history.columns = ['real_DAI_balance', 'real_ETH_balance', 'real_UNI_supply']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Despite the simplifications, the outputs of the model are very similar to the real data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fd77624a438>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,5))\n",
    "ax = plt.subplot(131)\n",
    "model_history.astype(float).plot(ax=ax, y=['model_ETH_balance'])\n",
    "real_history.astype(float).plot(ax=ax, y=['real_ETH_balance'])\n",
    "ax = plt.subplot(132)\n",
    "model_history.astype(float).plot(ax=ax, y=['model_DAI_balance'])\n",
    "real_history.astype(float).plot(ax=ax, y=['real_DAI_balance'])\n",
    "ax = plt.subplot(133)\n",
    "model_history.astype(float).plot(ax=ax, y=['model_UNI_supply'])\n",
    "real_history.astype(float).plot(ax=ax, y=['real_UNI_supply'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Only if we look much more closely can we see some deviation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>model_DAI_balance</th>\n",
       "      <th>model_ETH_balance</th>\n",
       "      <th>model_UNI_supply</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>147437</th>\n",
       "      <td>655398639698980968543553</td>\n",
       "      <td>5088972909434012930085</td>\n",
       "      <td>3308419107140103200194</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               model_DAI_balance       model_ETH_balance  \\\n",
       "147437  655398639698980968543553  5088972909434012930085   \n",
       "\n",
       "              model_UNI_supply  \n",
       "147437  3308419107140103200194  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_history.tail(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>real_DAI_balance</th>\n",
       "      <th>real_ETH_balance</th>\n",
       "      <th>real_UNI_supply</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>147437</th>\n",
       "      <td>662323537318822558984586</td>\n",
       "      <td>5009819142181678051370</td>\n",
       "      <td>3304234867026987131356</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                real_DAI_balance        real_ETH_balance  \\\n",
       "147437  662323537318822558984586  5009819142181678051370   \n",
       "\n",
       "               real_UNI_supply  \n",
       "147437  3304234867026987131356  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "real_history.tail(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9895445394438577"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_history.iloc[-1]['model_DAI_balance']/real_history.iloc[-1]['real_DAI_balance']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0157997255002433"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_history.iloc[-1]['model_ETH_balance']/real_history.iloc[-1]['real_ETH_balance']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.001266326481471"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_history.iloc[-1]['model_UNI_supply']/real_history.iloc[-1]['real_UNI_supply']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "\n",
    "In this notebook, we have illustrated how to create a cadCAD model that takes in real data and replicates the mechanics of a real world smart contract extremely acurately by translating the smart contract code into python code. We enforce best practices for cadCAD modeling, and analyze our simulation data against the real data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}