stocks_train_test.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Multi Stocks Strategy with multiple algorithms\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Crabs\\Anaconda3\\envs\\finrl37\\lib\\site-packages\\pyfolio\\pos.py:27: UserWarning: Module \"zipline.assets\" not found; multipliers will not be applied to position notionals.\n",
      "  'Module \"zipline.assets\" not found; multipliers will not be applied'\n"
     ]
    }
   ],
   "source": [
    "from config import stocks\n",
    "from config import general as config\n",
    "from finrl.meta.env_stock_trading.env_stocktrading import StockTradingEnv\n",
    "from lib.drl import load_dataset, data_split\n",
    "from lib.stocks import StocksStrategy\n",
    "from lib.support import check_directory_structure\n",
    "\n",
    "ROOT_DIR = '.'\n",
    "check_directory_structure(ROOT_DIR)\n",
    "\n",
    "FILE_PREFIX = \"stocks\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 99341 rows from file ./datasets/stocks/DOW30.csv\n",
      "Adding 8 indicators: macd, boll_ub, boll_lb, rsi_30, cci_30, dx_30, close_30_sma, close_60_sma\n",
      "Adding turbulence index\n",
      "Adding vix\n",
      "[*********************100%***********************]  1 of 1 completed\n",
      "Train Shape: (83897, 18)\n",
      "Test Shape: (14587, 18)\n"
     ]
    }
   ],
   "source": [
    "df = load_dataset(f'{ROOT_DIR}/datasets/stocks/DOW30.csv', stocks.INDICATORS, use_turbulence=True, use_vix=True)\n",
    "\n",
    "train_df = data_split(df, '2009-01-01', '2020-07-01')\n",
    "trade_df = data_split(df, '2020-07-01', '2022-07-01')\n",
    "\n",
    "print(f\"Train Shape: {train_df.shape}\")\n",
    "print(f\"Test Shape: {trade_df.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stock Dimension: 29, State Space: 291\n"
     ]
    }
   ],
   "source": [
    "stock_dimension = len(train_df.tic.unique())\n",
    "state_space = 1 + 2 * stock_dimension + len(stocks.INDICATORS) * stock_dimension\n",
    "print(f\"Stock Dimension: {stock_dimension}, State Space: {state_space}\")\n",
    "\n",
    "buy_cost_list = sell_cost_list = [0.001] * stock_dimension\n",
    "num_stock_shares = [0] * stock_dimension"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Train and Testing a single algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "env_kwargs = {\n",
    "    \"hmax\": 100,\n",
    "    \"initial_amount\": 1000000,\n",
    "    \"num_stock_shares\": num_stock_shares,\n",
    "    \"buy_cost_pct\": buy_cost_list,\n",
    "    \"sell_cost_pct\": sell_cost_list,\n",
    "    \"state_space\": state_space,\n",
    "    \"stock_dim\": stock_dimension,\n",
    "    \"tech_indicator_list\": stocks.INDICATORS,\n",
    "    \"action_space\": stock_dimension,\n",
    "    \"reward_scaling\": 1e-4\n",
    "}\n",
    "e_train_gym = StockTradingEnv(df=train_df, **env_kwargs)\n",
    "env_train, _ = e_train_gym.get_sb_env()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "stockStrategy = StocksStrategy(model_name=\"SAC\", root_dir=ROOT_DIR, results_dir=config.RESULTS_DIR, trained_model_dir=config.TRAINED_MODEL_DIR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# # TRAIN\n",
    "total_timesteps = 500\n",
    "stockStrategy.train(env_train, total_timesteps, use_existing=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "hit end!\n"
     ]
    }
   ],
   "source": [
    "# # TRAIN\n",
    "e_trade_gym = StockTradingEnv(df=trade_df, turbulence_threshold=70, risk_indicator_col='vix', **env_kwargs)\n",
    "stockStrategy.test(e_trade_gym)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training and Testing multiple algorithms at once"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_list = [\"SAC\", \"DDPG\", \"A2C\", \"PPO\", \"TD3\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SAC\n",
      "DDPG\n",
      "{'batch_size': 128, 'buffer_size': 50000, 'learning_rate': 0.001}\n",
      "Using cpu device\n",
      "A2C\n",
      "{'n_steps': 5, 'ent_coef': 0.01, 'learning_rate': 0.0007}\n",
      "Using cpu device\n",
      "PPO\n",
      "{'n_steps': 2048, 'ent_coef': 0.01, 'learning_rate': 0.00025, 'batch_size': 128}\n",
      "Using cpu device\n",
      "----------------------------------\n",
      "| time/              |           |\n",
      "|    fps             | 116       |\n",
      "|    iterations      | 1         |\n",
      "|    time_elapsed    | 17        |\n",
      "|    total_timesteps | 2048      |\n",
      "| train/             |           |\n",
      "|    reward          | 0.5371737 |\n",
      "----------------------------------\n",
      "TD3\n",
      "{'batch_size': 100, 'buffer_size': 1000000, 'learning_rate': 0.001}\n",
      "Using cpu device\n"
     ]
    }
   ],
   "source": [
    "for model_name in model_list:\n",
    "    stockStrategy = StocksStrategy(model_name, root_dir=ROOT_DIR, results_dir=config.RESULTS_DIR, trained_model_dir=config.TRAINED_MODEL_DIR)\n",
    "    print(model_name)\n",
    "    total_timesteps = 50 # 500\n",
    "    stockStrategy.train(env_train, total_timesteps, use_existing=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "hit end!\n",
      "hit end!\n",
      "hit end!\n",
      "hit end!\n",
      "hit end!\n"
     ]
    }
   ],
   "source": [
    "for model_name in model_list:\n",
    "    e_trade_gym = StockTradingEnv(df=trade_df, turbulence_threshold=70, risk_indicator_col='vix', **env_kwargs)\n",
    "    stockStrategy = StocksStrategy(model_name, root_dir=ROOT_DIR, results_dir=config.RESULTS_DIR, trained_model_dir=config.TRAINED_MODEL_DIR)\n",
    "    stockStrategy.test(e_trade_gym)"
   ]
  }
 ],
 "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": "3.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}