Baseline DQN Gym_Trading Tutorial.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import gym_trading\n",
    "import gym\n",
    "import sys\n",
    "import itertools\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import tensorflow.contrib.layers as layers\n",
    "\n",
    "import baselines.common.tf_util as U\n",
    "\n",
    "from baselines import deepq\n",
    "from baselines.deepq.replay_buffer import ReplayBuffer\n",
    "from baselines.common.schedules import LinearSchedule"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Trading Framework\n",
    "\n",
    "This framework is developed based on Peter Henry https://github.com/Henry-bee/gym_trading/ which in turn on developed of Tito Ingargiola's https://github.com/hackthemarket/gym-trading. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "First, define the address for the CSV data\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "csv = \"data/EURUSD60.csv\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create a new OpenAI Gym environment with the customised Trading environment\n",
    "\n",
    "\n",
    "\n",
    " .initialise_simulator() must be invoked after **env.make('trading-v0')** . Within this function, provide these arguments:\n",
    "\n",
    "\n",
    "* **csv_name**: Address of the data\n",
    "\n",
    "\n",
    "* **trade_period**: (int), Max of duration of each trades. *Default: 1000*\n",
    "\n",
    "\n",
    "* **train_split**: (float), Percentage of data set for training. *Default: 0.7*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[2017-07-09 18:39:23,128] Making new env: trading-v0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                       Return       ATR  Open Trade  Duration Trade\n",
      "Date_Time                                                          \n",
      "2013-12-02 02:00:00  0.421251  0.355142         0.0             0.0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/adrian/.local/lib/python3.5/site-packages/matplotlib/cbook.py:136: MatplotlibDeprecationWarning: The finance module has been deprecated in mpl 2.0 and will be removed in mpl 2.2. Please use the module mpl_finance instead.\n",
      "  warnings.warn(message, mplDeprecation, stacklevel=1)\n"
     ]
    }
   ],
   "source": [
    "env = gym.make('trading-v0')\n",
    "env.initialise_simulator(csv, trade_period=50, train_split=0.7)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# States map\n",
    "\n",
    "states_map is a discretized observation space bounded by the extreme values of features with an interval of 0.5. Also I use trade duration and boolean of active trade. This observations are dinamical, while the algorithm runs. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.42125144,  0.35514246,  0.        ,  0.        ],\n",
       "       [ 1.7094119 ,  0.45689009,  0.        ,  0.        ],\n",
       "       [-0.59190109,  0.45925152,  0.        ,  0.        ],\n",
       "       ..., \n",
       "       [-1.00049976, -0.45976391,  0.        ,  0.        ],\n",
       "       [-0.20124835, -0.51999441,  0.        ,  0.        ],\n",
       "       [ 0.31455146, -0.61636501,  0.        ,  0.        ]])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env.sim.states"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The magic (Deep Q-Network)\n",
    "\n",
    "The point of Baselines OpenAI is set of high-quality implementations of reinforcement learning algorithms. A lot of projects for reinforcement trading uses their own implementations, causing small bugs or hard to maintenance/improvement.\n",
    "\n",
    "There are a lot of good resources to drill down on this topic. \n",
    "But well above, the core of Q_learning and DQNs can express with the next diagrams\n",
    "\n",
    "![Image of DQN](https://raw.githubusercontent.com/udacity/deep-learning/master/reinforcement/assets/deep-q-learning.png)\n",
    "\n",
    "\n",
    "![Image of Q-Network](https://raw.githubusercontent.com/udacity/deep-learning/master/reinforcement/assets/q-network.png)\n",
    "\n",
    "\n",
    "### Learning resources:\n",
    "\n",
    "http://karpathy.github.io/2016/05/31/rl/\n",
    "\n",
    "http://minpy.readthedocs.io/en/latest/tutorial/rl_policy_gradient_tutorial/rl_policy_gradient.html\n",
    "\n",
    "http://pemami4911.github.io/blog/2016/08/21/ddpg-rl.html\n",
    "\n",
    "http://kvfrans.com/simple-algoritms-for-solving-cartpole/\n",
    "\n",
    "https://medium.com/@awjuliani/super-simple-reinforcement-learning-tutorial-part-1-fd544fab149\n",
    "\n",
    "https://dataorigami.net/blogs/napkin-folding/79031811-multi-armed-bandits\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Set the model \n",
    "\n",
    "So, let's get our hands dirty. First set our network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def model(inpt, num_actions, scope, reuse=False):\n",
    "    \"\"\"This model takes as input an observation and returns values of all actions.\"\"\"\n",
    "    with tf.variable_scope(scope, reuse=reuse):\n",
    "        out = inpt\n",
    "        out = layers.fully_connected(out, num_outputs=128, activation_fn=tf.nn.tanh)\n",
    "        out = layers.fully_connected(out, num_outputs=64, activation_fn=tf.nn.tanh)\n",
    "        out = layers.fully_connected(out, num_outputs=32, activation_fn=tf.nn.tanh)\n",
    "        out = layers.fully_connected(out, num_outputs=num_actions, activation_fn=None)\n",
    "        return out"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And define run_test function to use in the end of every episode"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def run_test(env, act, episodes=1, final_test=False):\n",
    "    obs = env._reset(train=False)\n",
    "    start = env.sim.train_end_index + 1\n",
    "    end = env.sim.count - 2\n",
    "\n",
    "    for episode in range(episodes):\n",
    "        done = False\n",
    "        while done is False:\n",
    "            action = act(obs[None])\n",
    "            obs, reward, done, info = env.step(action)\n",
    "\n",
    "        if not final_test:\n",
    "            journal = pd.DataFrame(env.portfolio.journal)\n",
    "            profit = journal[\"Profit\"].sum()\n",
    "            return env.portfolio.average_profit_per_trade, profit\n",
    "        else:\n",
    "            print(\"Training period  %s - %s\" % (env.sim.date_time[start], env.sim.date_time[end]))\n",
    "            print(\"Average Reward is %s\" % (env.portfolio.average_profit_per_trade))\n",
    "\n",
    "    if final_test:\n",
    "        env._generate_summary_stats()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Running the enviroment!\n",
    "\n",
    "At this point, we can start up the enviroment and run the episodes. The most important is:\n",
    "\n",
    "* Set the `episode_rewards` with the reward that we want. For example if we want maximice each trade: `episode_rewards[-1] += rew`\n",
    "* Set the solved function. The training will stop when the outcome of function get True. For example: `is_solved = np.mean(episode_rewards[-101:-1]) > 1000 or t == 100000`\n",
    "* Instanciate `deepq.build_train`. That is the core of Baseline of OpenAI.\n",
    "\n",
    "### `build_train` Creates the train function:\n",
    "\n",
    "#### Parameters\n",
    "\n",
    "* make_obs_ph: str -> tf.placeholder or TfInput -> a function that takes a name and creates a placeholder of input with that name\n",
    "\n",
    "* q_func: (tf.Variable, int, str, bool) -> tf.Variable -> the model that takes the following inputs:\n",
    "    * observation_in: object -> the output of observation placeholder\n",
    "    * num_actions: int -> number of actions\n",
    "    * scope: str\n",
    "    * reuse: bool -> should be passed to outer variable scope and returns a tensor of shape (batch_size, num_actions) with values of every action.\n",
    "\n",
    "* num_actions: int -> number of actions\n",
    "\n",
    "* reuse: bool -> whether or not to reuse the graph variables\n",
    "\n",
    "* optimizer: tf.train.Optimizer -> optimizer to use for the Q-learning objective.\n",
    "\n",
    "* grad_norm_clipping: float or None -> clip gradient norms to this value. If None no clipping is performed.\n",
    "\n",
    "* gamma: float -> discount rate.\n",
    "\n",
    "* double_q: bool -> if true will use Double Q Learning (https://arxiv.org/abs/1509.06461).In general it is a good idea to keep it enabled.\n",
    "\n",
    "* scope: str or VariableScope -> optional scope for variable_scope.\n",
    "\n",
    "* reuse: bool or None -> whether or not the variables should be reused. To be able to reuse the scope must be given.\n",
    "\n",
    "#### Returns\n",
    "\n",
    "* act: (tf.Variable, bool, float) -> tf.Variable -> function to select and action given observation.\n",
    "\n",
    "* train: (object, np.array, np.array, object, np.array, np.array) -> np.array  -> optimize the error in Bellman's equation.See the top of the file for details.\n",
    "\n",
    "* update_target: () -> () -> copy the parameters from optimized Q function to the target Q function.\n",
    "debug: {str: function} -> a bunch of functions to print debug data like q_values.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:VARIABLES collection name is deprecated, please use GLOBAL_VARIABLES instead; VARIABLES will be removed after 2017-03-02.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[2017-07-09 18:39:23,452] VARIABLES collection name is deprecated, please use GLOBAL_VARIABLES instead; VARIABLES will be removed after 2017-03-02.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:VARIABLES collection name is deprecated, please use GLOBAL_VARIABLES instead; VARIABLES will be removed after 2017-03-02.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[2017-07-09 18:39:23,474] VARIABLES collection name is deprecated, please use GLOBAL_VARIABLES instead; VARIABLES will be removed after 2017-03-02.\n",
      "/home/adrian/.local/lib/python3.5/site-packages/numpy/core/fromnumeric.py:2909: RuntimeWarning: Mean of empty slice.\n",
      "  out=out, **kwargs)\n",
      "/home/adrian/.local/lib/python3.5/site-packages/numpy/core/_methods.py:80: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  ret = ret.dtype.type(ret / rcount)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-------------------------------------\n",
      "steps                     | 2138\n",
      "episodes                  | 1\n",
      "% time spent exploring    | 79\n",
      "--\n",
      "mean episode reward       | nan\n",
      "Total operations          | 391\n",
      "Avg duration trades       | 3.07\n",
      "Total profit              | -960.0\n",
      "Avg profit per trade      | -2.872\n",
      "--\n",
      "Total profit test:        > -229.6\n",
      "Avg profit per trade test > -2.031\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 4277\n",
      "episodes                  | 2\n",
      "% time spent exploring    | 58\n",
      "--\n",
      "mean episode reward       | -4001.2\n",
      "Total operations          | 430\n",
      "Avg duration trades       | 2.02\n",
      "Total profit              | -1238.0\n",
      "Avg profit per trade      | -2.16\n",
      "--\n",
      "Total profit test:        > -313.4\n",
      "Avg profit per trade test > -2.581\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 6416\n",
      "episodes                  | 3\n",
      "% time spent exploring    | 37\n",
      "--\n",
      "mean episode reward       | -3412.6\n",
      "Total operations          | 379\n",
      "Avg duration trades       | 1.54\n",
      "Total profit              | -1294.0\n",
      "Avg profit per trade      | -3.681\n",
      "--\n",
      "Total profit test:        > -273.0\n",
      "Avg profit per trade test > -2.696\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 8555\n",
      "episodes                  | 4\n",
      "% time spent exploring    | 16\n",
      "--\n",
      "mean episode reward       | -3157.4\n",
      "Total operations          | 219\n",
      "Avg duration trades       | 1.32\n",
      "Total profit              | -998.0\n",
      "Avg profit per trade      | -4.135\n",
      "--\n",
      "Total profit test:        > -229.3\n",
      "Avg profit per trade test > -3.618\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 10694\n",
      "episodes                  | 5\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -2637.3\n",
      "Total operations          | 72\n",
      "Avg duration trades       | 1.08\n",
      "Total profit              | -359.0\n",
      "Avg profit per trade      | -4.692\n",
      "--\n",
      "Total profit test:        > -144.8\n",
      "Avg profit per trade test > -7.225\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 12833\n",
      "episodes                  | 6\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -2166.0\n",
      "Total operations          | 25\n",
      "Avg duration trades       | 1.04\n",
      "Total profit              | 2.0\n",
      "Avg profit per trade      | -2.008\n",
      "--\n",
      "Total profit test:        > -41.7\n",
      "Avg profit per trade test > -3.878\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 14972\n",
      "episodes                  | 7\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -1814.0\n",
      "Total operations          | 26\n",
      "Avg duration trades       | 1.04\n",
      "Total profit              | -73.0\n",
      "Avg profit per trade      | -2.515\n",
      "--\n",
      "Total profit test:        > -37.9\n",
      "Avg profit per trade test > -1.188\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 17111\n",
      "episodes                  | 8\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -1558.6\n",
      "Total operations          | 29\n",
      "Avg duration trades       | 1.1\n",
      "Total profit              | -165.0\n",
      "Avg profit per trade      | -5.676\n",
      "--\n",
      "Total profit test:        > -5.5\n",
      "Avg profit per trade test > -1.233\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 19250\n",
      "episodes                  | 9\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -1396.0\n",
      "Total operations          | 19\n",
      "Avg duration trades       | 1.0\n",
      "Total profit              | -61.0\n",
      "Avg profit per trade      | 0.189\n",
      "--\n",
      "Total profit test:        > -96.3\n",
      "Avg profit per trade test > -4.577\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 21389\n",
      "episodes                  | 10\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -1240.5\n",
      "Total operations          | 24\n",
      "Avg duration trades       | 1.04\n",
      "Total profit              | -30.0\n",
      "Avg profit per trade      | -3.242\n",
      "--\n",
      "Total profit test:        > -173.1\n",
      "Avg profit per trade test > -3.732\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 23528\n",
      "episodes                  | 11\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -1122.5\n",
      "Total operations          | 41\n",
      "Avg duration trades       | 1.05\n",
      "Total profit              | 25.0\n",
      "Avg profit per trade      | -0.298\n",
      "--\n",
      "Total profit test:        > -155.2\n",
      "Avg profit per trade test > -1.946\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 25667\n",
      "episodes                  | 12\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -1019.3\n",
      "Total operations          | 55\n",
      "Avg duration trades       | 1.11\n",
      "Total profit              | -259.0\n",
      "Avg profit per trade      | -3.622\n",
      "--\n",
      "Total profit test:        > -315.8\n",
      "Avg profit per trade test > -2.511\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 27806\n",
      "episodes                  | 13\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -951.7\n",
      "Total operations          | 33\n",
      "Avg duration trades       | 1.06\n",
      "Total profit              | -28.0\n",
      "Avg profit per trade      | -1.212\n",
      "--\n",
      "Total profit test:        > -64.3\n",
      "Avg profit per trade test > -1.829\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 29945\n",
      "episodes                  | 14\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -881.3\n",
      "Total operations          | 29\n",
      "Avg duration trades       | 1.1\n",
      "Total profit              | 159.0\n",
      "Avg profit per trade      | 1.659\n",
      "--\n",
      "Total profit test:        > -72.9\n",
      "Avg profit per trade test > -2.112\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 32084\n",
      "episodes                  | 15\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -816.4\n",
      "Total operations          | 47\n",
      "Avg duration trades       | 1.02\n",
      "Total profit              | -27.0\n",
      "Avg profit per trade      | -1.366\n",
      "--\n",
      "Total profit test:        > -48.8\n",
      "Avg profit per trade test > -2.744\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 34223\n",
      "episodes                  | 16\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -765.1\n",
      "Total operations          | 43\n",
      "Avg duration trades       | 1.09\n",
      "Total profit              | -10.0\n",
      "Avg profit per trade      | -0.905\n",
      "--\n",
      "Total profit test:        > -245.7\n",
      "Avg profit per trade test > -2.414\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 36362\n",
      "episodes                  | 17\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -720.5\n",
      "Total operations          | 42\n",
      "Avg duration trades       | 1.12\n",
      "Total profit              | -100.0\n",
      "Avg profit per trade      | -3.871\n",
      "--\n",
      "Total profit test:        > -72.0\n",
      "Avg profit per trade test > -0.417\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 38501\n",
      "episodes                  | 18\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -689.3\n",
      "Total operations          | 32\n",
      "Avg duration trades       | 1.03\n",
      "Total profit              | -3.0\n",
      "Avg profit per trade      | -0.553\n",
      "--\n",
      "Total profit test:        > -183.9\n",
      "Avg profit per trade test > -2.791\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 40640\n",
      "episodes                  | 19\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -651.0\n",
      "Total operations          | 24\n",
      "Avg duration trades       | 1.04\n",
      "Total profit              | 79.0\n",
      "Avg profit per trade      | 1.488\n",
      "--\n",
      "Total profit test:        > -99.8\n",
      "Avg profit per trade test > -1.883\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 42779\n",
      "episodes                  | 20\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -613.9\n",
      "Total operations          | 27\n",
      "Avg duration trades       | 1.07\n",
      "Total profit              | 65.0\n",
      "Avg profit per trade      | -0.622\n",
      "--\n",
      "Total profit test:        > -125.3\n",
      "Avg profit per trade test > -0.846\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 44918\n",
      "episodes                  | 21\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -583.0\n",
      "Total operations          | 34\n",
      "Avg duration trades       | 1.03\n",
      "Total profit              | -70.0\n",
      "Avg profit per trade      | -2.324\n",
      "--\n",
      "Total profit test:        > -50.9\n",
      "Avg profit per trade test > -1.518\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 47057\n",
      "episodes                  | 22\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -558.2\n",
      "Total operations          | 28\n",
      "Avg duration trades       | 1.07\n",
      "Total profit              | -57.0\n",
      "Avg profit per trade      | -3.571\n",
      "--\n",
      "Total profit test:        > -145.2\n",
      "Avg profit per trade test > -3.628\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 49196\n",
      "episodes                  | 23\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -538.3\n",
      "Total operations          | 21\n",
      "Avg duration trades       | 1.1\n",
      "Total profit              | 29.0\n",
      "Avg profit per trade      | -1.614\n",
      "--\n",
      "Total profit test:        > -111.5\n",
      "Avg profit per trade test > -3.698\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 51335\n",
      "episodes                  | 24\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -515.7\n",
      "Total operations          | 23\n",
      "Avg duration trades       | 1.04\n",
      "Total profit              | 23.0\n",
      "Avg profit per trade      | 2.03\n",
      "--\n",
      "Total profit test:        > 57.3\n",
      "Avg profit per trade test > -1.155\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 53474\n",
      "episodes                  | 25\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -491.6\n",
      "Total operations          | 24\n",
      "Avg duration trades       | 1.0\n",
      "Total profit              | 66.0\n",
      "Avg profit per trade      | 0.617\n",
      "--\n",
      "Total profit test:        > -32.7\n",
      "Avg profit per trade test > -1.377\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 55613\n",
      "episodes                  | 26\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -471.3\n",
      "Total operations          | 34\n",
      "Avg duration trades       | 1.12\n",
      "Total profit              | -7.0\n",
      "Avg profit per trade      | -2.132\n",
      "--\n",
      "Total profit test:        > -70.0\n",
      "Avg profit per trade test > -3.566\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 57752\n",
      "episodes                  | 27\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -456.0\n",
      "Total operations          | 34\n",
      "Avg duration trades       | 1.09\n",
      "Total profit              | -31.0\n",
      "Avg profit per trade      | -3.224\n",
      "--\n",
      "Total profit test:        > -197.5\n",
      "Avg profit per trade test > -2.964\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 59891\n",
      "episodes                  | 28\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -442.5\n",
      "Total operations          | 31\n",
      "Avg duration trades       | 1.03\n",
      "Total profit              | -29.0\n",
      "Avg profit per trade      | -1.029\n",
      "--\n",
      "Total profit test:        > -221.1\n",
      "Avg profit per trade test > -5.586\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 62030\n",
      "episodes                  | 29\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -427.2\n",
      "Total operations          | 30\n",
      "Avg duration trades       | 1.03\n",
      "Total profit              | 22.0\n",
      "Avg profit per trade      | 0.417\n",
      "--\n",
      "Total profit test:        > -81.5\n",
      "Avg profit per trade test > -1.489\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 64169\n",
      "episodes                  | 30\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -411.4\n",
      "Total operations          | 28\n",
      "Avg duration trades       | 1.07\n",
      "Total profit              | 91.0\n",
      "Avg profit per trade      | 0.029\n",
      "--\n",
      "Total profit test:        > -14.9\n",
      "Avg profit per trade test > -0.949\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 66308\n",
      "episodes                  | 31\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -397.3\n",
      "Total operations          | 32\n",
      "Avg duration trades       | 1.06\n",
      "Total profit              | -250.0\n",
      "Avg profit per trade      | -5.978\n",
      "--\n",
      "Total profit test:        > -109.1\n",
      "Avg profit per trade test > -3.825\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 68447\n",
      "episodes                  | 32\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -390.6\n",
      "Total operations          | 36\n",
      "Avg duration trades       | 1.06\n",
      "Total profit              | -14.0\n",
      "Avg profit per trade      | -2.475\n",
      "--\n",
      "Total profit test:        > -174.8\n",
      "Avg profit per trade test > -2.577\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 70586\n",
      "episodes                  | 33\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -380.6\n",
      "Total operations          | 32\n",
      "Avg duration trades       | 1.19\n",
      "Total profit              | 48.0\n",
      "Avg profit per trade      | -0.463\n",
      "--\n",
      "Total profit test:        > -139.3\n",
      "Avg profit per trade test > -1.747\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 72725\n",
      "episodes                  | 34\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -368.9\n",
      "Total operations          | 28\n",
      "Avg duration trades       | 1.11\n",
      "Total profit              | -1.0\n",
      "Avg profit per trade      | -3.196\n",
      "--\n",
      "Total profit test:        > -153.0\n",
      "Avg profit per trade test > -1.558\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 74864\n",
      "episodes                  | 35\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -360.2\n",
      "Total operations          | 27\n",
      "Avg duration trades       | 1.11\n",
      "Total profit              | 26.0\n",
      "Avg profit per trade      | 0.241\n",
      "--\n",
      "Total profit test:        > -172.7\n",
      "Avg profit per trade test > -2.689\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 77003\n",
      "episodes                  | 36\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -348.7\n",
      "Total operations          | 30\n",
      "Avg duration trades       | 1.07\n",
      "Total profit              | 23.0\n",
      "Avg profit per trade      | -0.703\n",
      "--\n",
      "Total profit test:        > -122.0\n",
      "Avg profit per trade test > -2.392\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 79142\n",
      "episodes                  | 37\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -338.6\n",
      "Total operations          | 34\n",
      "Avg duration trades       | 1.06\n",
      "Total profit              | 11.0\n",
      "Avg profit per trade      | 2.085\n",
      "--\n",
      "Total profit test:        > -135.7\n",
      "Avg profit per trade test > -6.794\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 81281\n",
      "episodes                  | 38\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -326.8\n",
      "Total operations          | 32\n",
      "Avg duration trades       | 1.06\n",
      "Total profit              | -72.0\n",
      "Avg profit per trade      | -2.744\n",
      "--\n",
      "Total profit test:        > -122.9\n",
      "Avg profit per trade test > -2.819\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 83420\n",
      "episodes                  | 39\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -319.6\n",
      "Total operations          | 21\n",
      "Avg duration trades       | 1.1\n",
      "Total profit              | -9.0\n",
      "Avg profit per trade      | -2.038\n",
      "--\n",
      "Total profit test:        > -122.6\n",
      "Avg profit per trade test > -1.194\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 85559\n",
      "episodes                  | 40\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -311.7\n",
      "Total operations          | 29\n",
      "Avg duration trades       | 1.21\n",
      "Total profit              | -62.0\n",
      "Avg profit per trade      | -0.838\n",
      "--\n",
      "Total profit test:        > -145.5\n",
      "Avg profit per trade test > -1.73\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 87698\n",
      "episodes                  | 41\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -303.0\n",
      "Total operations          | 33\n",
      "Avg duration trades       | 1.18\n",
      "Total profit              | 32.0\n",
      "Avg profit per trade      | -0.9\n",
      "--\n",
      "Total profit test:        > -173.6\n",
      "Avg profit per trade test > -2.388\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 89837\n",
      "episodes                  | 42\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -294.7\n",
      "Total operations          | 27\n",
      "Avg duration trades       | 1.33\n",
      "Total profit              | 142.0\n",
      "Avg profit per trade      | 2.481\n",
      "--\n",
      "Total profit test:        > -157.2\n",
      "Avg profit per trade test > -2.287\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 91976\n",
      "episodes                  | 43\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -283.4\n",
      "Total operations          | 19\n",
      "Avg duration trades       | 1.47\n",
      "Total profit              | 24.0\n",
      "Avg profit per trade      | 0.3\n",
      "--\n",
      "Total profit test:        > -212.0\n",
      "Avg profit per trade test > -2.696\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 94115\n",
      "episodes                  | 44\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -274.5\n",
      "Total operations          | 27\n",
      "Avg duration trades       | 1.7\n",
      "Total profit              | 135.0\n",
      "Avg profit per trade      | 2.086\n",
      "--\n",
      "Total profit test:        > -134.9\n",
      "Avg profit per trade test > -2.287\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 96254\n",
      "episodes                  | 45\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -263.6\n",
      "Total operations          | 39\n",
      "Avg duration trades       | 2.36\n",
      "Total profit              | 105.0\n",
      "Avg profit per trade      | 1.702\n",
      "--\n",
      "Total profit test:        > -221.5\n",
      "Avg profit per trade test > -2.88\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 98393\n",
      "episodes                  | 46\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -264.1\n",
      "Total operations          | 51\n",
      "Avg duration trades       | 2.16\n",
      "Total profit              | 127.0\n",
      "Avg profit per trade      | 1.357\n",
      "--\n",
      "Total profit test:        > -124.3\n",
      "Avg profit per trade test > -2.08\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 100532\n",
      "episodes                  | 47\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -256.6\n",
      "Total operations          | 53\n",
      "Avg duration trades       | 2.53\n",
      "Total profit              | 132.0\n",
      "Avg profit per trade      | 2.434\n",
      "--\n",
      "Total profit test:        > -175.4\n",
      "Avg profit per trade test > -4.422\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 102671\n",
      "episodes                  | 48\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -232.8\n",
      "Total operations          | 67\n",
      "Avg duration trades       | 4.81\n",
      "Total profit              | 53.0\n",
      "Avg profit per trade      | -0.725\n",
      "--\n",
      "Total profit test:        > -371.3\n",
      "Avg profit per trade test > -4.285\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 104810\n",
      "episodes                  | 49\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -253.5\n",
      "Total operations          | 50\n",
      "Avg duration trades       | 4.76\n",
      "Total profit              | 171.0\n",
      "Avg profit per trade      | 2.255\n",
      "--\n",
      "Total profit test:        > -358.2\n",
      "Avg profit per trade test > -7.422\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 106949\n",
      "episodes                  | 50\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -200.4\n",
      "Total operations          | 78\n",
      "Avg duration trades       | 6.01\n",
      "Total profit              | -184.0\n",
      "Avg profit per trade      | -2.765\n",
      "--\n",
      "Total profit test:        > 141.1\n",
      "Avg profit per trade test > 1.08\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 109088\n",
      "episodes                  | 51\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -261.9\n",
      "Total operations          | 74\n",
      "Avg duration trades       | 5.3\n",
      "Total profit              | -209.0\n",
      "Avg profit per trade      | -2.991\n",
      "--\n",
      "Total profit test:        > 299.2\n",
      "Avg profit per trade test > 7.845\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 111227\n",
      "episodes                  | 52\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -293.0\n",
      "Total operations          | 96\n",
      "Avg duration trades       | 4.18\n",
      "Total profit              | -474.0\n",
      "Avg profit per trade      | -2.953\n",
      "--\n",
      "Total profit test:        > -222.5\n",
      "Avg profit per trade test > -3.047\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 113366\n",
      "episodes                  | 53\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -310.3\n",
      "Total operations          | 91\n",
      "Avg duration trades       | 3.82\n",
      "Total profit              | -401.0\n",
      "Avg profit per trade      | -1.372\n",
      "--\n",
      "Total profit test:        > -424.0\n",
      "Avg profit per trade test > -5.69\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 115505\n",
      "episodes                  | 54\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -303.5\n",
      "Total operations          | 51\n",
      "Avg duration trades       | 2.84\n",
      "Total profit              | 34.0\n",
      "Avg profit per trade      | -0.114\n",
      "--\n",
      "Total profit test:        > -273.3\n",
      "Avg profit per trade test > -3.617\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 117644\n",
      "episodes                  | 55\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -293.7\n",
      "Total operations          | 64\n",
      "Avg duration trades       | 3.73\n",
      "Total profit              | -248.0\n",
      "Avg profit per trade      | -1.477\n",
      "--\n",
      "Total profit test:        > -203.2\n",
      "Avg profit per trade test > -4.718\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 119783\n",
      "episodes                  | 56\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -282.2\n",
      "Total operations          | 61\n",
      "Avg duration trades       | 4.02\n",
      "Total profit              | 45.0\n",
      "Avg profit per trade      | -0.679\n",
      "--\n",
      "Total profit test:        > -367.9\n",
      "Avg profit per trade test > -4.455\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 121922\n",
      "episodes                  | 57\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -258.2\n",
      "Total operations          | 58\n",
      "Avg duration trades       | 3.81\n",
      "Total profit              | -229.0\n",
      "Avg profit per trade      | -2.321\n",
      "--\n",
      "Total profit test:        > -365.9\n",
      "Avg profit per trade test > -3.31\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 124061\n",
      "episodes                  | 58\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -234.9\n",
      "Total operations          | 60\n",
      "Avg duration trades       | 4.17\n",
      "Total profit              | -481.0\n",
      "Avg profit per trade      | -6.095\n",
      "--\n",
      "Total profit test:        > -67.2\n",
      "Avg profit per trade test > -2.657\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 126200\n",
      "episodes                  | 59\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -241.9\n",
      "Total operations          | 91\n",
      "Avg duration trades       | 3.89\n",
      "Total profit              | -559.0\n",
      "Avg profit per trade      | -2.791\n",
      "--\n",
      "Total profit test:        > -524.3\n",
      "Avg profit per trade test > -4.596\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 128339\n",
      "episodes                  | 60\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -247.4\n",
      "Total operations          | 35\n",
      "Avg duration trades       | 4.37\n",
      "Total profit              | 76.0\n",
      "Avg profit per trade      | 2.12\n",
      "--\n",
      "Total profit test:        > -328.1\n",
      "Avg profit per trade test > -3.563\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 130478\n",
      "episodes                  | 61\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -229.6\n",
      "Total operations          | 68\n",
      "Avg duration trades       | 5.29\n",
      "Total profit              | -471.0\n",
      "Avg profit per trade      | -4.554\n",
      "--\n",
      "Total profit test:        > -294.4\n",
      "Avg profit per trade test > -3.465\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 132617\n",
      "episodes                  | 62\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -220.3\n",
      "Total operations          | 56\n",
      "Avg duration trades       | 5.16\n",
      "Total profit              | 114.0\n",
      "Avg profit per trade      | 2.179\n",
      "--\n",
      "Total profit test:        > -372.8\n",
      "Avg profit per trade test > -4.898\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 134756\n",
      "episodes                  | 63\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -178.8\n",
      "Total operations          | 88\n",
      "Avg duration trades       | 5.34\n",
      "Total profit              | 30.0\n",
      "Avg profit per trade      | -0.115\n",
      "--\n",
      "Total profit test:        > -114.8\n",
      "Avg profit per trade test > -2.296\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 136895\n",
      "episodes                  | 64\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -176.3\n",
      "Total operations          | 129\n",
      "Avg duration trades       | 5.79\n",
      "Total profit              | -895.0\n",
      "Avg profit per trade      | -6.125\n",
      "--\n",
      "Total profit test:        > -396.8\n",
      "Avg profit per trade test > -3.712\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 139034\n",
      "episodes                  | 65\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -220.5\n",
      "Total operations          | 142\n",
      "Avg duration trades       | 4.72\n",
      "Total profit              | -976.0\n",
      "Avg profit per trade      | -4.858\n",
      "--\n",
      "Total profit test:        > -512.4\n",
      "Avg profit per trade test > -7.022\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 141173\n",
      "episodes                  | 66\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -271.9\n",
      "Total operations          | 66\n",
      "Avg duration trades       | 5.26\n",
      "Total profit              | -277.0\n",
      "Avg profit per trade      | -2.55\n",
      "--\n",
      "Total profit test:        > -554.5\n",
      "Avg profit per trade test > -6.777\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 143312\n",
      "episodes                  | 67\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -286.9\n",
      "Total operations          | 50\n",
      "Avg duration trades       | 5.18\n",
      "Total profit              | 92.0\n",
      "Avg profit per trade      | 2.398\n",
      "--\n",
      "Total profit test:        > -352.5\n",
      "Avg profit per trade test > -3.916\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 145451\n",
      "episodes                  | 68\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -258.0\n",
      "Total operations          | 52\n",
      "Avg duration trades       | 5.63\n",
      "Total profit              | -100.0\n",
      "Avg profit per trade      | -2.238\n",
      "--\n",
      "Total profit test:        > -414.7\n",
      "Avg profit per trade test > -5.205\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 147590\n",
      "episodes                  | 69\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -266.3\n",
      "Total operations          | 63\n",
      "Avg duration trades       | 3.21\n",
      "Total profit              | -30.0\n",
      "Avg profit per trade      | -0.91\n",
      "--\n",
      "Total profit test:        > -406.4\n",
      "Avg profit per trade test > -2.836\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 149729\n",
      "episodes                  | 70\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -256.6\n",
      "Total operations          | 50\n",
      "Avg duration trades       | 5.96\n",
      "Total profit              | 34.0\n",
      "Avg profit per trade      | 0.657\n",
      "--\n",
      "Total profit test:        > -519.4\n",
      "Avg profit per trade test > -5.599\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 151868\n",
      "episodes                  | 71\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -219.0\n",
      "Total operations          | 80\n",
      "Avg duration trades       | 5.54\n",
      "Total profit              | -7.0\n",
      "Avg profit per trade      | 2.668\n",
      "--\n",
      "Total profit test:        > -289.8\n",
      "Avg profit per trade test > -3.255\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 154007\n",
      "episodes                  | 72\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -204.6\n",
      "Total operations          | 75\n",
      "Avg duration trades       | 6.12\n",
      "Total profit              | -411.0\n",
      "Avg profit per trade      | -2.493\n",
      "--\n",
      "Total profit test:        > -506.0\n",
      "Avg profit per trade test > -5.549\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 156146\n",
      "episodes                  | 73\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -212.1\n",
      "Total operations          | 65\n",
      "Avg duration trades       | 4.82\n",
      "Total profit              | -141.0\n",
      "Avg profit per trade      | -1.011\n",
      "--\n",
      "Total profit test:        > -354.4\n",
      "Avg profit per trade test > -2.282\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 158285\n",
      "episodes                  | 74\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -181.1\n",
      "Total operations          | 60\n",
      "Avg duration trades       | 4.85\n",
      "Total profit              | 170.0\n",
      "Avg profit per trade      | 1.239\n",
      "--\n",
      "Total profit test:        > -225.0\n",
      "Avg profit per trade test > -1.956\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 160424\n",
      "episodes                  | 75\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -146.1\n",
      "Total operations          | 60\n",
      "Avg duration trades       | 5.6\n",
      "Total profit              | -279.0\n",
      "Avg profit per trade      | -1.56\n",
      "--\n",
      "Total profit test:        > -463.5\n",
      "Avg profit per trade test > -6.091\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 162563\n",
      "episodes                  | 76\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -156.0\n",
      "Total operations          | 68\n",
      "Avg duration trades       | 7.79\n",
      "Total profit              | -278.0\n",
      "Avg profit per trade      | -2.712\n",
      "--\n",
      "Total profit test:        > -432.1\n",
      "Avg profit per trade test > -4.569\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 164702\n",
      "episodes                  | 77\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -167.8\n",
      "Total operations          | 73\n",
      "Avg duration trades       | 6.4\n",
      "Total profit              | -585.0\n",
      "Avg profit per trade      | -7.718\n",
      "--\n",
      "Total profit test:        > -354.5\n",
      "Avg profit per trade test > -4.427\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 166841\n",
      "episodes                  | 78\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -183.0\n",
      "Total operations          | 56\n",
      "Avg duration trades       | 4.71\n",
      "Total profit              | -92.0\n",
      "Avg profit per trade      | -1.081\n",
      "--\n",
      "Total profit test:        > -397.4\n",
      "Avg profit per trade test > -6.212\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 168980\n",
      "episodes                  | 79\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -152.8\n",
      "Total operations          | 38\n",
      "Avg duration trades       | 13.32\n",
      "Total profit              | 170.0\n",
      "Avg profit per trade      | 5.454\n",
      "--\n",
      "Total profit test:        > -204.6\n",
      "Avg profit per trade test > -9.535\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 171119\n",
      "episodes                  | 80\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -96.2\n",
      "Total operations          | 56\n",
      "Avg duration trades       | 8.29\n",
      "Total profit              | -33.0\n",
      "Avg profit per trade      | -1.126\n",
      "--\n",
      "Total profit test:        > -145.2\n",
      "Avg profit per trade test > -2.958\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 173258\n",
      "episodes                  | 81\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -118.4\n",
      "Total operations          | 66\n",
      "Avg duration trades       | 7.18\n",
      "Total profit              | -228.0\n",
      "Avg profit per trade      | -1.469\n",
      "--\n",
      "Total profit test:        > -611.2\n",
      "Avg profit per trade test > -10.065\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 175397\n",
      "episodes                  | 82\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -104.4\n",
      "Total operations          | 54\n",
      "Avg duration trades       | 8.48\n",
      "Total profit              | -159.0\n",
      "Avg profit per trade      | 1.967\n",
      "--\n",
      "Total profit test:        > -282.9\n",
      "Avg profit per trade test > -6.31\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 177536\n",
      "episodes                  | 83\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -142.8\n",
      "Total operations          | 60\n",
      "Avg duration trades       | 7.53\n",
      "Total profit              | -157.0\n",
      "Avg profit per trade      | -0.392\n",
      "--\n",
      "Total profit test:        > -361.0\n",
      "Avg profit per trade test > -8.871\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 179675\n",
      "episodes                  | 84\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -99.9\n",
      "Total operations          | 77\n",
      "Avg duration trades       | 5.45\n",
      "Total profit              | -487.0\n",
      "Avg profit per trade      | -2.562\n",
      "--\n",
      "Total profit test:        > -245.2\n",
      "Avg profit per trade test > -2.668\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 181814\n",
      "episodes                  | 85\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -92.5\n",
      "Total operations          | 77\n",
      "Avg duration trades       | 4.95\n",
      "Total profit              | -156.0\n",
      "Avg profit per trade      | -4.25\n",
      "--\n",
      "Total profit test:        > -476.2\n",
      "Avg profit per trade test > -3.188\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 183953\n",
      "episodes                  | 86\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -150.9\n",
      "Total operations          | 64\n",
      "Avg duration trades       | 5.09\n",
      "Total profit              | 111.0\n",
      "Avg profit per trade      | -1.071\n",
      "--\n",
      "Total profit test:        > -493.7\n",
      "Avg profit per trade test > -4.503\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 186092\n",
      "episodes                  | 87\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -94.2\n",
      "Total operations          | 62\n",
      "Avg duration trades       | 5.32\n",
      "Total profit              | 27.0\n",
      "Avg profit per trade      | -1.087\n",
      "--\n",
      "Total profit test:        > -505.7\n",
      "Avg profit per trade test > -4.711\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 188231\n",
      "episodes                  | 88\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -109.1\n",
      "Total operations          | 68\n",
      "Avg duration trades       | 5.43\n",
      "Total profit              | 204.0\n",
      "Avg profit per trade      | -1.159\n",
      "--\n",
      "Total profit test:        > -631.5\n",
      "Avg profit per trade test > -7.235\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 190370\n",
      "episodes                  | 89\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -114.1\n",
      "Total operations          | 72\n",
      "Avg duration trades       | 5.89\n",
      "Total profit              | -377.0\n",
      "Avg profit per trade      | -3.36\n",
      "--\n",
      "Total profit test:        > -323.7\n",
      "Avg profit per trade test > -5.596\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 192509\n",
      "episodes                  | 90\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -146.4\n",
      "Total operations          | 83\n",
      "Avg duration trades       | 4.89\n",
      "Total profit              | 58.0\n",
      "Avg profit per trade      | -0.707\n",
      "--\n",
      "Total profit test:        > -499.6\n",
      "Avg profit per trade test > -4.705\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 194648\n",
      "episodes                  | 91\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -121.9\n",
      "Total operations          | 73\n",
      "Avg duration trades       | 6.47\n",
      "Total profit              | 176.0\n",
      "Avg profit per trade      | 1.499\n",
      "--\n",
      "Total profit test:        > -574.2\n",
      "Avg profit per trade test > -5.747\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 196787\n",
      "episodes                  | 92\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -93.4\n",
      "Total operations          | 65\n",
      "Avg duration trades       | 6.28\n",
      "Total profit              | 494.0\n",
      "Avg profit per trade      | 5.395\n",
      "--\n",
      "Total profit test:        > -615.5\n",
      "Avg profit per trade test > -6.767\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 198926\n",
      "episodes                  | 93\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 8.8\n",
      "Total operations          | 69\n",
      "Avg duration trades       | 3.91\n",
      "Total profit              | 147.0\n",
      "Avg profit per trade      | -0.51\n",
      "--\n",
      "Total profit test:        > -411.7\n",
      "Avg profit per trade test > -4.868\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 201065\n",
      "episodes                  | 94\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 13.2\n",
      "Total operations          | 68\n",
      "Avg duration trades       | 4.18\n",
      "Total profit              | -183.0\n",
      "Avg profit per trade      | -3.897\n",
      "--\n",
      "Total profit test:        > -307.2\n",
      "Avg profit per trade test > -3.095\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 203204\n",
      "episodes                  | 95\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -18.2\n",
      "Total operations          | 70\n",
      "Avg duration trades       | 3.19\n",
      "Total profit              | -136.0\n",
      "Avg profit per trade      | -3.918\n",
      "--\n",
      "Total profit test:        > -502.9\n",
      "Avg profit per trade test > -4.275\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 205343\n",
      "episodes                  | 96\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -8.7\n",
      "Total operations          | 50\n",
      "Avg duration trades       | 4.4\n",
      "Total profit              | 113.0\n",
      "Avg profit per trade      | 2.527\n",
      "--\n",
      "Total profit test:        > -523.1\n",
      "Avg profit per trade test > -5.099\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 207482\n",
      "episodes                  | 97\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -16.3\n",
      "Total operations          | 68\n",
      "Avg duration trades       | 4.57\n",
      "Total profit              | -88.0\n",
      "Avg profit per trade      | -1.681\n",
      "--\n",
      "Total profit test:        > -506.0\n",
      "Avg profit per trade test > -7.191\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 209621\n",
      "episodes                  | 98\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -16.4\n",
      "Total operations          | 88\n",
      "Avg duration trades       | 4.59\n",
      "Total profit              | -145.0\n",
      "Avg profit per trade      | -1.27\n",
      "--\n",
      "Total profit test:        > -281.2\n",
      "Avg profit per trade test > -6.603\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 211760\n",
      "episodes                  | 99\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -43.4\n",
      "Total operations          | 86\n",
      "Avg duration trades       | 5.03\n",
      "Total profit              | -219.0\n",
      "Avg profit per trade      | -2.174\n",
      "--\n",
      "Total profit test:        > 26.1\n",
      "Avg profit per trade test > -0.3\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 213899\n",
      "episodes                  | 100\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -65.7\n",
      "Total operations          | 128\n",
      "Avg duration trades       | 4.28\n",
      "Total profit              | -363.0\n",
      "Avg profit per trade      | -3.468\n",
      "--\n",
      "Total profit test:        > -459.2\n",
      "Avg profit per trade test > -11.926\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 216038\n",
      "episodes                  | 101\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -105.6\n",
      "Total operations          | 74\n",
      "Avg duration trades       | 5.36\n",
      "Total profit              | -57.0\n",
      "Avg profit per trade      | 0.227\n",
      "--\n",
      "Total profit test:        > -238.1\n",
      "Avg profit per trade test > -4.857\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 218177\n",
      "episodes                  | 102\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -71.5\n",
      "Total operations          | 58\n",
      "Avg duration trades       | 5.26\n",
      "Total profit              | 62.0\n",
      "Avg profit per trade      | -0.003\n",
      "--\n",
      "Total profit test:        > -249.0\n",
      "Avg profit per trade test > -4.298\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 220316\n",
      "episodes                  | 103\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -61.2\n",
      "Total operations          | 41\n",
      "Avg duration trades       | 5.78\n",
      "Total profit              | -286.0\n",
      "Avg profit per trade      | -1.055\n",
      "--\n",
      "Total profit test:        > -505.5\n",
      "Avg profit per trade test > -9.154\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 222455\n",
      "episodes                  | 104\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -23.8\n",
      "Total operations          | 44\n",
      "Avg duration trades       | 5.09\n",
      "Total profit              | 24.0\n",
      "Avg profit per trade      | -0.862\n",
      "--\n",
      "Total profit test:        > -495.1\n",
      "Avg profit per trade test > -5.536\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 224594\n",
      "episodes                  | 105\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -6.8\n",
      "Total operations          | 43\n",
      "Avg duration trades       | 4.79\n",
      "Total profit              | -88.0\n",
      "Avg profit per trade      | -3.695\n",
      "--\n",
      "Total profit test:        > -75.2\n",
      "Avg profit per trade test > -3.815\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 226733\n",
      "episodes                  | 106\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -32.1\n",
      "Total operations          | 59\n",
      "Avg duration trades       | 1.86\n",
      "Total profit              | -92.0\n",
      "Avg profit per trade      | -1.058\n",
      "--\n",
      "Total profit test:        > -426.2\n",
      "Avg profit per trade test > -5.631\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 228872\n",
      "episodes                  | 107\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -28.8\n",
      "Total operations          | 45\n",
      "Avg duration trades       | 1.93\n",
      "Total profit              | 66.0\n",
      "Avg profit per trade      | -2.98\n",
      "--\n",
      "Total profit test:        > -526.1\n",
      "Avg profit per trade test > -6.055\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 231011\n",
      "episodes                  | 108\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -27.6\n",
      "Total operations          | 63\n",
      "Avg duration trades       | 1.65\n",
      "Total profit              | 111.0\n",
      "Avg profit per trade      | 0.381\n",
      "--\n",
      "Total profit test:        > -486.5\n",
      "Avg profit per trade test > -4.157\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 233150\n",
      "episodes                  | 109\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -22.3\n",
      "Total operations          | 61\n",
      "Avg duration trades       | 2.03\n",
      "Total profit              | 234.0\n",
      "Avg profit per trade      | 0.556\n",
      "--\n",
      "Total profit test:        > -341.6\n",
      "Avg profit per trade test > -3.423\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 235289\n",
      "episodes                  | 110\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -12.9\n",
      "Total operations          | 71\n",
      "Avg duration trades       | 2.27\n",
      "Total profit              | -70.0\n",
      "Avg profit per trade      | -1.342\n",
      "--\n",
      "Total profit test:        > -501.7\n",
      "Avg profit per trade test > -4.464\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 237428\n",
      "episodes                  | 111\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -35.1\n",
      "Total operations          | 59\n",
      "Avg duration trades       | 2.03\n",
      "Total profit              | 321.0\n",
      "Avg profit per trade      | 1.241\n",
      "--\n",
      "Total profit test:        > -490.6\n",
      "Avg profit per trade test > -4.865\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 239567\n",
      "episodes                  | 112\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -31.6\n",
      "Total operations          | 59\n",
      "Avg duration trades       | 1.86\n",
      "Total profit              | 123.0\n",
      "Avg profit per trade      | -0.507\n",
      "--\n",
      "Total profit test:        > -377.9\n",
      "Avg profit per trade test > -4.192\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 241706\n",
      "episodes                  | 113\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -29.4\n",
      "Total operations          | 51\n",
      "Avg duration trades       | 1.47\n",
      "Total profit              | 51.0\n",
      "Avg profit per trade      | 0.443\n",
      "--\n",
      "Total profit test:        > -407.6\n",
      "Avg profit per trade test > -4.718\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 243845\n",
      "episodes                  | 114\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -28.0\n",
      "Total operations          | 50\n",
      "Avg duration trades       | 1.88\n",
      "Total profit              | 127.0\n",
      "Avg profit per trade      | -0.14\n",
      "--\n",
      "Total profit test:        > -455.9\n",
      "Avg profit per trade test > -4.828\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 245984\n",
      "episodes                  | 115\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -24.1\n",
      "Total operations          | 63\n",
      "Avg duration trades       | 1.75\n",
      "Total profit              | 111.0\n",
      "Avg profit per trade      | 0.305\n",
      "--\n",
      "Total profit test:        > -509.0\n",
      "Avg profit per trade test > -4.954\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 248123\n",
      "episodes                  | 116\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -18.5\n",
      "Total operations          | 68\n",
      "Avg duration trades       | 1.84\n",
      "Total profit              | -5.0\n",
      "Avg profit per trade      | -0.29\n",
      "--\n",
      "Total profit test:        > -388.5\n",
      "Avg profit per trade test > -4.913\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 250262\n",
      "episodes                  | 117\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -15.0\n",
      "Total operations          | 53\n",
      "Avg duration trades       | 2.04\n",
      "Total profit              | 234.0\n",
      "Avg profit per trade      | 0.96\n",
      "--\n",
      "Total profit test:        > -397.4\n",
      "Avg profit per trade test > -4.829\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 252401\n",
      "episodes                  | 118\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -7.4\n",
      "Total operations          | 92\n",
      "Avg duration trades       | 1.64\n",
      "Total profit              | 10.0\n",
      "Avg profit per trade      | -0.49\n",
      "--\n",
      "Total profit test:        > -297.7\n",
      "Avg profit per trade test > -3.096\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 254540\n",
      "episodes                  | 119\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -4.4\n",
      "Total operations          | 61\n",
      "Avg duration trades       | 2.33\n",
      "Total profit              | 95.0\n",
      "Avg profit per trade      | -0.405\n",
      "--\n",
      "Total profit test:        > -311.8\n",
      "Avg profit per trade test > -3.635\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 256679\n",
      "episodes                  | 120\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | -1.9\n",
      "Total operations          | 54\n",
      "Avg duration trades       | 2.02\n",
      "Total profit              | 29.0\n",
      "Avg profit per trade      | 0.624\n",
      "--\n",
      "Total profit test:        > -533.3\n",
      "Avg profit per trade test > -4.602\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 258818\n",
      "episodes                  | 121\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 0.8\n",
      "Total operations          | 57\n",
      "Avg duration trades       | 2.11\n",
      "Total profit              | 129.0\n",
      "Avg profit per trade      | 0.526\n",
      "--\n",
      "Total profit test:        > -484.1\n",
      "Avg profit per trade test > -4.915\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 260957\n",
      "episodes                  | 122\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 6.0\n",
      "Total operations          | 46\n",
      "Avg duration trades       | 3.7\n",
      "Total profit              | 84.0\n",
      "Avg profit per trade      | -0.413\n",
      "--\n",
      "Total profit test:        > -552.4\n",
      "Avg profit per trade test > -5.417\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 263096\n",
      "episodes                  | 123\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 11.1\n",
      "Total operations          | 45\n",
      "Avg duration trades       | 5.47\n",
      "Total profit              | 4.0\n",
      "Avg profit per trade      | 1.591\n",
      "--\n",
      "Total profit test:        > -184.0\n",
      "Avg profit per trade test > -2.802\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 265235\n",
      "episodes                  | 124\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 21.3\n",
      "Total operations          | 95\n",
      "Avg duration trades       | 3.56\n",
      "Total profit              | 99.0\n",
      "Avg profit per trade      | 2.021\n",
      "--\n",
      "Total profit test:        > -210.6\n",
      "Avg profit per trade test > -5.125\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 267374\n",
      "episodes                  | 125\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 59.2\n",
      "Total operations          | 56\n",
      "Avg duration trades       | 7.07\n",
      "Total profit              | -25.0\n",
      "Avg profit per trade      | 1.204\n",
      "--\n",
      "Total profit test:        > -181.1\n",
      "Avg profit per trade test > -5.492\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 269513\n",
      "episodes                  | 126\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 72.8\n",
      "Total operations          | 105\n",
      "Avg duration trades       | 7.2\n",
      "Total profit              | -71.0\n",
      "Avg profit per trade      | 0.523\n",
      "--\n",
      "Total profit test:        > -308.8\n",
      "Avg profit per trade test > -13.9\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 271652\n",
      "episodes                  | 127\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 128.3\n",
      "Total operations          | 153\n",
      "Avg duration trades       | 7.65\n",
      "Total profit              | -313.0\n",
      "Avg profit per trade      | -0.681\n",
      "--\n",
      "Total profit test:        > -132.7\n",
      "Avg profit per trade test > -4.605\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 273791\n",
      "episodes                  | 128\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 134.5\n",
      "Total operations          | 84\n",
      "Avg duration trades       | 12.81\n",
      "Total profit              | -38.0\n",
      "Avg profit per trade      | 0.785\n",
      "--\n",
      "Total profit test:        > -279.7\n",
      "Avg profit per trade test > -9.3\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 275930\n",
      "episodes                  | 129\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 191.9\n",
      "Total operations          | 93\n",
      "Avg duration trades       | 11.56\n",
      "Total profit              | 96.0\n",
      "Avg profit per trade      | -1.676\n",
      "--\n",
      "Total profit test:        > -207.7\n",
      "Avg profit per trade test > -8.28\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 278069\n",
      "episodes                  | 130\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 212.9\n",
      "Total operations          | 90\n",
      "Avg duration trades       | 15.38\n",
      "Total profit              | -115.0\n",
      "Avg profit per trade      | 0.568\n",
      "--\n",
      "Total profit test:        > -177.2\n",
      "Avg profit per trade test > -5.861\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 280208\n",
      "episodes                  | 131\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 197.6\n",
      "Total operations          | 114\n",
      "Avg duration trades       | 14.01\n",
      "Total profit              | -202.0\n",
      "Avg profit per trade      | -0.016\n",
      "--\n",
      "Total profit test:        > -68.7\n",
      "Avg profit per trade test > -2.85\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 282347\n",
      "episodes                  | 132\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 277.4\n",
      "Total operations          | 111\n",
      "Avg duration trades       | 15.96\n",
      "Total profit              | 237.0\n",
      "Avg profit per trade      | 2.396\n",
      "--\n",
      "Total profit test:        > -207.3\n",
      "Avg profit per trade test > -9.785\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 284486\n",
      "episodes                  | 133\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 266.1\n",
      "Total operations          | 96\n",
      "Avg duration trades       | 19.39\n",
      "Total profit              | -218.0\n",
      "Avg profit per trade      | -0.936\n",
      "--\n",
      "Total profit test:        > -237.8\n",
      "Avg profit per trade test > -9.595\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 286625\n",
      "episodes                  | 134\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 232.5\n",
      "Total operations          | 72\n",
      "Avg duration trades       | 26.15\n",
      "Total profit              | -124.0\n",
      "Avg profit per trade      | 0.777\n",
      "--\n",
      "Total profit test:        > -247.2\n",
      "Avg profit per trade test > -9.252\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 288764\n",
      "episodes                  | 135\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 118.4\n",
      "Total operations          | 81\n",
      "Avg duration trades       | 22.62\n",
      "Total profit              | -388.0\n",
      "Avg profit per trade      | -2.056\n",
      "--\n",
      "Total profit test:        > -263.7\n",
      "Avg profit per trade test > -9.942\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 290903\n",
      "episodes                  | 136\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 96.8\n",
      "Total operations          | 61\n",
      "Avg duration trades       | 32.85\n",
      "Total profit              | -570.0\n",
      "Avg profit per trade      | -7.089\n",
      "--\n",
      "Total profit test:        > -249.3\n",
      "Avg profit per trade test > -12.182\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 293042\n",
      "episodes                  | 137\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 21.4\n",
      "Total operations          | 55\n",
      "Avg duration trades       | 36.8\n",
      "Total profit              | 491.0\n",
      "Avg profit per trade      | 9.164\n",
      "--\n",
      "Total profit test:        > -261.1\n",
      "Avg profit per trade test > -11.267\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 295181\n",
      "episodes                  | 138\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 168.8\n",
      "Total operations          | 59\n",
      "Avg duration trades       | 34.03\n",
      "Total profit              | -98.0\n",
      "Avg profit per trade      | -0.388\n",
      "--\n",
      "Total profit test:        > -250.0\n",
      "Avg profit per trade test > -13.862\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 297320\n",
      "episodes                  | 139\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 130.1\n",
      "Total operations          | 74\n",
      "Avg duration trades       | 27.39\n",
      "Total profit              | 112.0\n",
      "Avg profit per trade      | 2.003\n",
      "--\n",
      "Total profit test:        > -249.9\n",
      "Avg profit per trade test > -9.914\n",
      "-------------------------------------\n",
      "-------------------------------------\n",
      "steps                     | 299459\n",
      "episodes                  | 140\n",
      "% time spent exploring    | 2\n",
      "--\n",
      "mean episode reward       | 197.2\n",
      "Total operations          | 93\n",
      "Avg duration trades       | 21.28\n",
      "Total profit              | 99.0\n",
      "Avg profit per trade      | 1.421\n",
      "--\n",
      "Total profit test:        > -260.8\n",
      "Avg profit per trade test > -11.521\n",
      "-------------------------------------\n",
      "SUMMARY STATISTICS\n",
      "Total Trades Taken:  14\n",
      "Total Reward:  347.3\n",
      "Average Reward per Trade:  24.8071428571\n",
      "Win Ratio: 71.4285714286 %\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACOcAAAJCCAYAAACBeTERAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f5cf41bdc88>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ { 'Entry Price': 1.3588100000000001,\n",
      "    'Entry Time': Timestamp('2013-12-02 03:00:00'),\n",
      "    'Exit Price': 1.35856,\n",
      "    'Exit Time': Timestamp('2013-12-04 05:00:00'),\n",
      "    'Profit': -5.5000000000008349,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -2.1000000000007546},\n",
      "  { 'Entry Price': 1.35856,\n",
      "    'Entry Time': Timestamp('2013-12-04 06:00:00'),\n",
      "    'Exit Price': 1.3654899999999999,\n",
      "    'Exit Time': Timestamp('2013-12-06 08:00:00'),\n",
      "    'Profit': 66.299999999998803,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 70.600000000000364},\n",
      "  { 'Entry Price': 1.3654999999999999,\n",
      "    'Entry Time': Timestamp('2013-12-06 09:00:00'),\n",
      "    'Exit Price': 1.37453,\n",
      "    'Exit Time': Timestamp('2013-12-10 11:00:00'),\n",
      "    'Profit': 87.300000000000935,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 94.400000000000801},\n",
      "  { 'Entry Price': 1.3745100000000001,\n",
      "    'Entry Time': Timestamp('2013-12-10 12:00:00'),\n",
      "    'Exit Price': 1.3781600000000001,\n",
      "    'Exit Time': Timestamp('2013-12-12 14:00:00'),\n",
      "    'Profit': 33.499999999999311,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 38.499999999998749},\n",
      "  { 'Entry Price': 1.37819,\n",
      "    'Entry Time': Timestamp('2013-12-12 15:00:00'),\n",
      "    'Exit Price': 1.3752899999999999,\n",
      "    'Exit Time': Timestamp('2013-12-16 17:00:00'),\n",
      "    'Profit': -32.000000000001251,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -25.600000000001497},\n",
      "  { 'Entry Price': 1.3752799999999998,\n",
      "    'Entry Time': Timestamp('2013-12-16 18:00:00'),\n",
      "    'Exit Price': 1.3746,\n",
      "    'Exit Time': Timestamp('2013-12-16 19:00:00'),\n",
      "    'Profit': -9.7999999999979188,\n",
      "    'Trade Duration': 1,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -0.099999999996544631},\n",
      "  { 'Entry Price': 1.3745799999999999,\n",
      "    'Entry Time': Timestamp('2013-12-16 20:00:00'),\n",
      "    'Exit Price': 1.3693200000000001,\n",
      "    'Exit Time': Timestamp('2013-12-18 22:00:00'),\n",
      "    'Profit': -55.599999999998204,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 7.4000000000015191},\n",
      "  { 'Entry Price': 1.3693,\n",
      "    'Entry Time': Timestamp('2013-12-18 23:00:00'),\n",
      "    'Exit Price': 1.3670599999999999,\n",
      "    'Exit Time': Timestamp('2013-12-19 15:00:00'),\n",
      "    'Profit': 19.400000000000198,\n",
      "    'Trade Duration': 16,\n",
      "    'Type': 'SELL',\n",
      "    'reward': 28.200000000000109},\n",
      "  { 'Entry Price': 1.36707,\n",
      "    'Entry Time': Timestamp('2013-12-19 16:00:00'),\n",
      "    'Exit Price': 1.36646,\n",
      "    'Exit Time': Timestamp('2013-12-19 18:00:00'),\n",
      "    'Profit': 3.0999999999999943,\n",
      "    'Trade Duration': 2,\n",
      "    'Type': 'SELL',\n",
      "    'reward': 0.29999999999941451},\n",
      "  { 'Entry Price': 1.3659299999999999,\n",
      "    'Entry Time': Timestamp('2013-12-19 21:00:00'),\n",
      "    'Exit Price': 1.3672,\n",
      "    'Exit Time': Timestamp('2013-12-20 18:00:00'),\n",
      "    'Profit': 9.7000000000010438,\n",
      "    'Trade Duration': 21,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 8.3000000000007574},\n",
      "  { 'Entry Price': 1.3670799999999999,\n",
      "    'Entry Time': Timestamp('2013-12-20 21:00:00'),\n",
      "    'Exit Price': 1.36842,\n",
      "    'Exit Time': Timestamp('2013-12-26 09:00:00'),\n",
      "    'Profit': 10.400000000001189,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 10.700000000000934},\n",
      "  { 'Entry Price': 1.3684100000000001,\n",
      "    'Entry Time': Timestamp('2013-12-26 10:00:00'),\n",
      "    'Exit Price': 1.3762700000000001,\n",
      "    'Exit Time': Timestamp('2013-12-30 12:00:00'),\n",
      "    'Profit': 75.599999999999781,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 64.599999999998801},\n",
      "  { 'Entry Price': 1.3762799999999999,\n",
      "    'Entry Time': Timestamp('2013-12-30 13:00:00'),\n",
      "    'Exit Price': 1.3668799999999999,\n",
      "    'Exit Time': Timestamp('2014-01-02 18:00:00'),\n",
      "    'Profit': 91.000000000000753,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'SELL',\n",
      "    'reward': 102.59999999999904},\n",
      "  { 'Entry Price': 1.3668799999999999,\n",
      "    'Entry Time': Timestamp('2014-01-02 19:00:00'),\n",
      "    'Exit Price': 1.3611899999999999,\n",
      "    'Exit Time': Timestamp('2014-01-03 17:00:00'),\n",
      "    'Profit': 53.899999999999729,\n",
      "    'Trade Duration': 22,\n",
      "    'Type': 'SELL',\n",
      "    'reward': 57.199999999999143}]\n",
      "Training period  2014-04-08 20:00:00 - 2014-05-30 22:00:00\n",
      "Average Reward is 3.775\n",
      "SUMMARY STATISTICS\n",
      "Total Trades Taken:  19\n",
      "Total Reward:  33.1\n",
      "Average Reward per Trade:  1.74210526316\n",
      "Win Ratio: 42.1052631579 %\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACOcAAAJCCAYAAACBeTERAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAACACfn/rUItSV7DOfMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f5cf41bdef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ { 'Entry Price': 1.3794999999999999,\n",
      "    'Entry Time': Timestamp('2014-04-08 20:00:00'),\n",
      "    'Exit Price': 1.3884100000000001,\n",
      "    'Exit Time': Timestamp('2014-04-10 22:00:00'),\n",
      "    'Profit': 86.100000000001955,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 91.400000000001143},\n",
      "  { 'Entry Price': 1.38839,\n",
      "    'Entry Time': Timestamp('2014-04-10 23:00:00'),\n",
      "    'Exit Price': 1.3815600000000001,\n",
      "    'Exit Time': Timestamp('2014-04-15 01:00:00'),\n",
      "    'Profit': -71.299999999998917,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -69.500000000000455},\n",
      "  { 'Entry Price': 1.3815600000000001,\n",
      "    'Entry Time': Timestamp('2014-04-15 02:00:00'),\n",
      "    'Exit Price': 1.3835999999999999,\n",
      "    'Exit Time': Timestamp('2014-04-17 04:00:00'),\n",
      "    'Profit': 17.399999999998197,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 18.599999999999397},\n",
      "  { 'Entry Price': 1.3835999999999999,\n",
      "    'Entry Time': Timestamp('2014-04-17 05:00:00'),\n",
      "    'Exit Price': 1.3815299999999999,\n",
      "    'Exit Time': Timestamp('2014-04-21 07:00:00'),\n",
      "    'Profit': -23.700000000000163,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -29.200000000000667},\n",
      "  { 'Entry Price': 1.38168,\n",
      "    'Entry Time': Timestamp('2014-04-21 09:00:00'),\n",
      "    'Exit Price': 1.38462,\n",
      "    'Exit Time': Timestamp('2014-04-23 11:00:00'),\n",
      "    'Profit': 26.399999999999427,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 0.99999999999955946},\n",
      "  { 'Entry Price': 1.3847499999999999,\n",
      "    'Entry Time': Timestamp('2014-04-23 12:00:00'),\n",
      "    'Exit Price': 1.38154,\n",
      "    'Exit Time': Timestamp('2014-04-24 00:00:00'),\n",
      "    'Profit': -35.099999999999355,\n",
      "    'Trade Duration': 12,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -35.20000000000001},\n",
      "  { 'Entry Price': 1.3815200000000001,\n",
      "    'Entry Time': Timestamp('2014-04-24 01:00:00'),\n",
      "    'Exit Price': 1.38195,\n",
      "    'Exit Time': Timestamp('2014-04-28 03:00:00'),\n",
      "    'Profit': 1.2999999999993044,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 19.899999999999032},\n",
      "  { 'Entry Price': 1.3819399999999999,\n",
      "    'Entry Time': Timestamp('2014-04-28 04:00:00'),\n",
      "    'Exit Price': 1.38052,\n",
      "    'Exit Time': Timestamp('2014-04-30 06:00:00'),\n",
      "    'Profit': -17.199999999999768,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -16.799999999999347},\n",
      "  { 'Entry Price': 1.3805399999999999,\n",
      "    'Entry Time': Timestamp('2014-04-30 07:00:00'),\n",
      "    'Exit Price': 1.3856299999999999,\n",
      "    'Exit Time': Timestamp('2014-05-02 09:00:00'),\n",
      "    'Profit': 47.900000000000389,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 50.200000000002127},\n",
      "  { 'Entry Price': 1.3856299999999999,\n",
      "    'Entry Time': Timestamp('2014-05-02 10:00:00'),\n",
      "    'Exit Price': 1.3924399999999999,\n",
      "    'Exit Time': Timestamp('2014-05-06 12:00:00'),\n",
      "    'Profit': 65.099999999999824,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 65.099999999999838},\n",
      "  { 'Entry Price': 1.39246,\n",
      "    'Entry Time': Timestamp('2014-05-06 13:00:00'),\n",
      "    'Exit Price': 1.3909,\n",
      "    'Exit Time': Timestamp('2014-05-08 15:00:00'),\n",
      "    'Profit': -18.600000000000058,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 18.999999999999797},\n",
      "  { 'Entry Price': 1.3862299999999999,\n",
      "    'Entry Time': Timestamp('2014-05-08 18:00:00'),\n",
      "    'Exit Price': 1.37575,\n",
      "    'Exit Time': Timestamp('2014-05-12 20:00:00'),\n",
      "    'Profit': 101.79999999999822,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'SELL',\n",
      "    'reward': 103.19999999999852},\n",
      "  { 'Entry Price': 1.37574,\n",
      "    'Entry Time': Timestamp('2014-05-12 21:00:00'),\n",
      "    'Exit Price': 1.37137,\n",
      "    'Exit Time': Timestamp('2014-05-14 12:00:00'),\n",
      "    'Profit': -46.699999999999847,\n",
      "    'Trade Duration': 39,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -36.199999999999896},\n",
      "  { 'Entry Price': 1.37137,\n",
      "    'Entry Time': Timestamp('2014-05-14 13:00:00'),\n",
      "    'Exit Price': 1.3704100000000001,\n",
      "    'Exit Time': Timestamp('2014-05-16 15:00:00'),\n",
      "    'Profit': -12.599999999998499,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -17.799999999999265},\n",
      "  { 'Entry Price': 1.3704499999999999,\n",
      "    'Entry Time': Timestamp('2014-05-16 16:00:00'),\n",
      "    'Exit Price': 1.36999,\n",
      "    'Exit Time': Timestamp('2014-05-20 18:00:00'),\n",
      "    'Profit': -7.5999999999990493,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -12.899999999998244},\n",
      "  { 'Entry Price': 1.37002,\n",
      "    'Entry Time': Timestamp('2014-05-20 19:00:00'),\n",
      "    'Exit Price': 1.3649799999999999,\n",
      "    'Exit Time': Timestamp('2014-05-22 21:00:00'),\n",
      "    'Profit': -53.400000000001555,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -57.600000000000207},\n",
      "  { 'Entry Price': 1.36496,\n",
      "    'Entry Time': Timestamp('2014-05-22 22:00:00'),\n",
      "    'Exit Price': 1.3638999999999999,\n",
      "    'Exit Time': Timestamp('2014-05-26 15:00:00'),\n",
      "    'Profit': -13.600000000000609,\n",
      "    'Trade Duration': 41,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -13.200000000000212},\n",
      "  { 'Entry Price': 1.3638999999999999,\n",
      "    'Entry Time': Timestamp('2014-05-26 16:00:00'),\n",
      "    'Exit Price': 1.35934,\n",
      "    'Exit Time': Timestamp('2014-05-28 18:00:00'),\n",
      "    'Profit': -48.599999999998971,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': -47.399999999997775},\n",
      "  { 'Entry Price': 1.3593600000000001,\n",
      "    'Entry Time': Timestamp('2014-05-28 19:00:00'),\n",
      "    'Exit Price': 1.36321,\n",
      "    'Exit Time': Timestamp('2014-05-30 21:00:00'),\n",
      "    'Profit': 35.499999999999091,\n",
      "    'Trade Duration': 50,\n",
      "    'Type': 'BUY',\n",
      "    'reward': 44.999999999999162}]\n"
     ]
    }
   ],
   "source": [
    "with U.make_session(8):\n",
    "    \n",
    "    act, train, update_target, debug = deepq.build_train(\n",
    "        make_obs_ph=lambda name: U.BatchInput(env.observation_space.shape, name=name),\n",
    "        q_func=model,\n",
    "        num_actions=env.action_space.n,\n",
    "        optimizer=tf.train.AdamOptimizer(learning_rate=5e-4),\n",
    "    )\n",
    "\n",
    "    replay_buffer = ReplayBuffer(50000)\n",
    "    # Create the schedule for exploration starting from 1 (every action is random) down to\n",
    "    # 0.02 (98% of actions are selected according to values predicted by the model).\n",
    "    exploration = LinearSchedule(schedule_timesteps=10000, initial_p=1.0, final_p=0.02)\n",
    "    # Initialize the parameters and copy them to the target network.\n",
    "    U.initialize()\n",
    "    update_target()\n",
    "\n",
    "    episode_rewards = [0.0]\n",
    "    obs = env.reset()\n",
    "    l_mean_episode_reward = []\n",
    "    for t in itertools.count():\n",
    "        # Take action and update exploration to the newest value\n",
    "        action = act(obs[None], update_eps=exploration.value(t))[0]\n",
    "\n",
    "        new_obs, rew, done, _ = env.step(action)\n",
    "\n",
    "        # Store transition in the replay buffer.\n",
    "        replay_buffer.add(obs, action, rew, new_obs, float(done))\n",
    "\n",
    "        obs = new_obs\n",
    "\n",
    "        episode_rewards[-1] += rew\n",
    "\n",
    "        is_solved = np.mean(episode_rewards[-101:-1]) > 500 or t >= 300000\n",
    "        is_solved = is_solved and len(env.portfolio.journal) > 2\n",
    "        \n",
    "        if done:\n",
    "\n",
    "            journal = pd.DataFrame(env.portfolio.journal)\n",
    "            profit = journal[\"Profit\"].sum()\n",
    "\n",
    "            try:\n",
    "                print(\"-------------------------------------\")\n",
    "                print(\"steps                     | {:}\".format(t))\n",
    "                print(\"episodes                  | {}\".format(len(episode_rewards)))\n",
    "                print(\"% time spent exploring    | {}\".format(int(100 * exploration.value(t))))\n",
    "\n",
    "                print(\"--\")\n",
    "                l_mean_episode_reward.append(round(np.mean(episode_rewards[-101:-1]), 1))\n",
    "\n",
    "                print(\"mean episode reward       | {:}\".format(l_mean_episode_reward[-1]))\n",
    "                print(\"Total operations          | {}\".format(len(env.portfolio.journal)))\n",
    "                print(\"Avg duration trades       | {}\".format(round(journal[\"Trade Duration\"].mean(), 2)))\n",
    "                print(\"Total profit              | {}\".format(round(profit), 1))\n",
    "                print(\"Avg profit per trade      | {}\".format(round(env.portfolio.average_profit_per_trade, 3)))\n",
    "\n",
    "                print(\"--\")\n",
    "\n",
    "                reward_test, profit = run_test(env=env, act=act)\n",
    "                print(\"Total profit test:        > {}\".format(round(profit, 2)))\n",
    "                print(\"Avg profit per trade test > {}\".format(round(reward_test, 3)))\n",
    "                print(\"-------------------------------------\")\n",
    "            except Exception as e:\n",
    "                print(\"Exception: \", e)\n",
    "                # Update target network periodically.\n",
    "\n",
    "            obs = env.reset()\n",
    "            episode_rewards.append(0)\n",
    "\n",
    "\n",
    "\n",
    "        if is_solved:\n",
    "            # Show off the result\n",
    "            env._generate_summary_stats()\n",
    "            run_test(env, act, final_test=True)\n",
    "            break\n",
    "\n",
    "        else:\n",
    "            # Minimize the error in Bellman's equation on a batch sampled from replay buffer.\n",
    "            if t > 500:\n",
    "                obses_t, actions, rewards, obses_tp1, dones = replay_buffer.sample(32)\n",
    "                train(obses_t, actions, rewards, obses_tp1, dones, np.ones_like(rewards))\n",
    "            if t % 500 == 0:\n",
    "                update_target()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZcAAAEKCAYAAADenhiQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f5cb00b29e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.plot(l_mean_episode_reward)\n",
    "plt.xlabel('Episode')\n",
    "plt.ylabel('Total Reward')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### It converges!\n",
    "Time to implement your own strategy :) \n",
    "\n",
    "And remember, be careful of overfitting.\n"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}