bottleneck.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-29T16:11:31.179451",
     "start_time": "2017-05-29T16:11:31.133955Z"
    },
    "code_folding": [],
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The line_profiler extension is already loaded. To reload it, use:\n",
      "  %reload_ext line_profiler\n"
     ]
    }
   ],
   "source": [
    "# -*- coding: utf-8 -*-\n",
    "CUDA_VISIBLE_DEVICES=-1 \n",
    "import os\n",
    "import pandas as pd\n",
    "from multiprocessing import Pool\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "import numpy as np\n",
    "import scipy\n",
    "\n",
    "import pybacktest as pb\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import threading\n",
    "import multiprocessing\n",
    "import tensorflow as tf\n",
    "import tensorflow.contrib.slim as slim\n",
    "import scipy.signal\n",
    "%matplotlib inline\n",
    "from random import choice\n",
    "from time import sleep\n",
    "from time import time\n",
    "import sys\n",
    "from trader_gym import environment\n",
    "# from A3C_class import *\n",
    "import mpld3\n",
    "mpld3.enable_notebook()\n",
    "# %load_ext ipycache\n",
    "max_episode_length = 300\n",
    "from profilehooks import profile\n",
    "\n",
    "%load_ext line_profiler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-29T16:14:20.237936",
     "start_time": "2017-05-29T16:14:18.629366Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# -*- coding: utf-8 -*-\n",
    "EXTRA_DENSE = False\n",
    "N_HIDDEN = 64\n",
    "DROPOUT = True\n",
    "COOL_V = True\n",
    "dep = 1\n",
    "# Тесты\n",
    "gamma = .8\n",
    "#магическая константа 37 - число фич\n",
    "s_size = 38*dep\n",
    "a_size = 3\n",
    "model_path = 'model'\n",
    "\n",
    "import os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import scipy\n",
    "\n",
    "import tensorflow as tf\n",
    "import tensorflow.contrib.slim as slim\n",
    "# %matplotlib inline\n",
    "from random import choice\n",
    "from trader_gym import environment\n",
    "# %load_ext ipycache\n",
    "\n",
    "\n",
    "def update_target_graph(from_scope,to_scope):\n",
    "    from_vars = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, from_scope)\n",
    "    to_vars = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, to_scope)\n",
    "\n",
    "    op_holder = []\n",
    "    for from_var,to_var in zip(from_vars,to_vars):\n",
    "        op_holder.append(to_var.assign(from_var))\n",
    "    return op_holder\n",
    "\n",
    "def discount(x, gamma):\n",
    "    return scipy.signal.lfilter([1], [1, -gamma], x[::-1], axis=0)[::-1]\n",
    "\n",
    "def normalized_columns_initializer(std=1.0):\n",
    "    def _initializer(shape, dtype=None, partition_info=None):\n",
    "        out = np.random.randn(*shape).astype(np.float32)\n",
    "        out *= std / np.sqrt(np.square(out).sum(axis=0, keepdims=True))\n",
    "        return tf.constant(out)\n",
    "    return _initializer\n",
    "\n",
    "class AC_Network():\n",
    "    def __init__(self,s_size,a_size,scope,trainer):\n",
    "        with tf.variable_scope(scope):\n",
    "            self.inputs = tf.placeholder(shape=[None,s_size],dtype=tf.float32)\n",
    "            self.imageIn = tf.reshape(self.inputs,shape=[-1,s_size])\n",
    "            if(EXTRA_DENSE == True):\n",
    "                hidden = slim.fully_connected(slim.flatten(self.imageIn),N_HIDDEN,activation_fn=tf.nn.tanh)\n",
    "            else:\n",
    "                hidden = slim.flatten(self.imageIn)\n",
    "            \n",
    "            if(DROPOUT == True):\n",
    "                rnn_in = tf.layers.dropout(hidden, rate=0.5, noise_shape=None, seed=None, training=False, name='drop1')\n",
    "            else:\n",
    "                rnn_in = hidden\n",
    "            \n",
    "            lstm_cell =  tf.contrib.rnn.BasicLSTMCell(N_HIDDEN,state_is_tuple=True)\n",
    "            c_init = np.zeros((1, lstm_cell.state_size.c), np.float32)\n",
    "            h_init = np.zeros((1, lstm_cell.state_size.h), np.float32)\n",
    "            self.state_init = [c_init, h_init]\n",
    "            c_in = tf.placeholder(tf.float32, [1, lstm_cell.state_size.c])\n",
    "            h_in = tf.placeholder(tf.float32, [1, lstm_cell.state_size.h])\n",
    "            self.state_in = (c_in, h_in)\n",
    "            rnn_in = tf.expand_dims(rnn_in, [0])\n",
    "            step_size = tf.shape(self.imageIn)[:1]\n",
    "            state_in = tf.contrib.rnn.LSTMStateTuple(c_in, h_in)\n",
    "            lstm_outputs, lstm_state = tf.nn.dynamic_rnn(\n",
    "                lstm_cell, rnn_in, initial_state=state_in, sequence_length=step_size,\n",
    "                time_major=False)\n",
    "            lstm_c, lstm_h = lstm_state\n",
    "            self.state_out = (lstm_c[:1, :], lstm_h[:1, :])\n",
    "            rnn_out = tf.reshape(lstm_outputs, [-1, N_HIDDEN])\n",
    "            \n",
    "            self.policy = slim.fully_connected(rnn_out,a_size,\n",
    "                activation_fn=tf.nn.softmax,\n",
    "                weights_initializer=normalized_columns_initializer(0.01),\n",
    "                biases_initializer=None)\n",
    "            if(COOL_V):\n",
    "                v_in = slim.fully_connected(slim.flatten(rnn_out),32,activation_fn=tf.nn.tanh)\n",
    "            else:\n",
    "                v_in = rnn_out\n",
    "                \n",
    "            self.value = slim.fully_connected(v_in,1,\n",
    "                activation_fn=None,\n",
    "                weights_initializer=normalized_columns_initializer(0.01),\n",
    "                biases_initializer=None)\n",
    "            \n",
    "            if scope != 'global':\n",
    "                self.actions = tf.placeholder(shape=[None],dtype=tf.int32)\n",
    "                self.actions_onehot = tf.one_hot(self.actions,a_size,dtype=tf.float32)\n",
    "                self.target_v = tf.placeholder(shape=[None],dtype=tf.float32)\n",
    "                self.advantages = tf.placeholder(shape=[None],dtype=tf.float32)\n",
    "                self.responsible_outputs = tf.reduce_sum(self.policy * self.actions_onehot, [1])\n",
    "\n",
    "                self.value_loss = 0.5 * tf.reduce_sum(tf.square(self.target_v - tf.reshape(self.value,[-1])))\n",
    "                self.entropy = - tf.reduce_sum(self.policy * tf.log(self.policy))\n",
    "                self.policy_loss = -tf.reduce_sum(tf.log(self.responsible_outputs)*self.advantages)\n",
    "                self.loss = 0.5 * self.value_loss + self.policy_loss - self.entropy * 0.01\n",
    "\n",
    "                local_vars = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, scope)\n",
    "                self.gradients = tf.gradients(self.loss,local_vars)\n",
    "                self.var_norms = tf.global_norm(local_vars)\n",
    "                grads,self.grad_norms = tf.clip_by_global_norm(self.gradients,40.0)\n",
    "                \n",
    "                global_vars = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, 'global')\n",
    "                self.apply_grads = trainer.apply_gradients(zip(grads,global_vars))\n",
    "\n",
    "class Worker():\n",
    "    def __init__(self, env, name,s_size,a_size,trainer,model_path,global_episodes):\n",
    "        self.name = \"worker_\" + str(name)\n",
    "        self.number = name        \n",
    "        self.model_path = model_path\n",
    "        self.trainer = trainer\n",
    "        self.global_episodes = global_episodes\n",
    "        self.increment = self.global_episodes.assign_add(1)\n",
    "        self.episode_rewards = []\n",
    "        self.episode_lengths = []\n",
    "        self.episode_mean_values = []\n",
    "        self.summary_writer = tf.summary.FileWriter('tb/train_'+str(self.number))\n",
    "\n",
    "        self.local_AC = AC_Network(s_size,a_size,self.name,trainer)\n",
    "        self.update_local_ops = update_target_graph('global',self.name)        \n",
    "        self.env = env\n",
    "        self.actions = [-1, 0, 1]\n",
    "           \n",
    "    def train(self,rollout,sess,gamma,bootstrap_value):\n",
    "        rollout = np.array(rollout)\n",
    "        observations = rollout[:,0]\n",
    "        actions = rollout[:,1]\n",
    "        rewards = rollout[:,2]\n",
    "        next_observations = rollout[:,3]\n",
    "        values = rollout[:,5]\n",
    "        \n",
    "        self.rewards_plus = np.asarray(rewards.tolist() + [bootstrap_value])\n",
    "        discounted_rewards = discount(self.rewards_plus,gamma)[:-1]\n",
    "        self.value_plus = np.asarray(values.tolist() + [bootstrap_value])\n",
    "        advantages = rewards + gamma * self.value_plus[1:] - self.value_plus[:-1]\n",
    "        advantages = discount(advantages,gamma)\n",
    "        #лол, это равно discounted_rewaeds - sefl.value_plus\n",
    "        \n",
    "        rnn_state = self.local_AC.state_init\n",
    "        feed_dict = {self.local_AC.target_v:discounted_rewards,\n",
    "            self.local_AC.inputs:np.vstack(observations),\n",
    "            self.local_AC.actions:actions,\n",
    "            self.local_AC.advantages:advantages,\n",
    "            self.local_AC.state_in[0]:rnn_state[0],\n",
    "            self.local_AC.state_in[1]:rnn_state[1]}\n",
    "        v_l,p_l,e_l,g_n,v_n,_ = sess.run([self.local_AC.value_loss,\n",
    "            self.local_AC.policy_loss,\n",
    "            self.local_AC.entropy,\n",
    "            self.local_AC.grad_norms,\n",
    "            self.local_AC.var_norms,\n",
    "            self.local_AC.apply_grads],\n",
    "            feed_dict=feed_dict)\n",
    "        return v_l / len(rollout),p_l / len(rollout),e_l / len(rollout), g_n,v_n\n",
    "    \n",
    "    \n",
    "    def work(self,max_episode_length,gamma,sess,saver, dep):\n",
    "        episode_count = sess.run(self.global_episodes)\n",
    "        total_steps = 0\n",
    "        print (\"Starting worker \" + str(self.number))\n",
    "        with sess.as_default(), sess.graph.as_default():    \n",
    "            while not coord.should_stop():\n",
    "                sess.run(self.update_local_ops)\n",
    "                episode_buffer = []\n",
    "                episode_values = []\n",
    "                episode_frames = []\n",
    "                action_buffer = [0]*dep\n",
    "                episode_reward = 0\n",
    "                episode_step_count = 0\n",
    "                d = False\n",
    "                s = self.env.reset()\n",
    "                s = np.concatenate((s, action_buffer))\n",
    "                episode_frames.append(s)\n",
    "                rnn_state = self.local_AC.state_init\n",
    "                summary = tf.Summary()\n",
    "\n",
    "                while d == False:\n",
    "                    a_dist,v,rnn_state = sess.run([self.local_AC.policy,self.local_AC.value,self.local_AC.state_out], \n",
    "                        feed_dict={self.local_AC.inputs:[s],\n",
    "                        self.local_AC.state_in[0]:rnn_state[0],\n",
    "                        self.local_AC.state_in[1]:rnn_state[1]})\n",
    "                    a = np.random.choice(a_dist[0],p=a_dist[0])\n",
    "                    a = np.argmax(a_dist == a)\n",
    "                    s1, r, d, _ = self.env.step(self.actions[a])\n",
    "                    #Добавим новое действие в конец буфера\n",
    "                    action_buffer[:] = np.concatenate((action_buffer[1:],[self.actions[a]]))\n",
    "                    s1 = np.concatenate((s1, action_buffer))\n",
    "                    if d == False:\n",
    "                        episode_frames.append(s1)\n",
    "                    else:\n",
    "                        s1 = s\n",
    "\n",
    "                    episode_buffer.append([s,a,r,s1,d,v[0,0]])\n",
    "                    episode_values.append(v[0,0])\n",
    "                    episode_reward += r\n",
    "                    s = s1      \n",
    "                    total_steps += 1\n",
    "                    episode_step_count += 1\n",
    "#                     Save history of boss actions\n",
    "\n",
    "                    if len(episode_buffer) == 200 and d != True and episode_step_count != max_episode_length - 1:\n",
    "                        v1 = sess.run(self.local_AC.value, \n",
    "                            feed_dict={self.local_AC.inputs:[s],\n",
    "                            self.local_AC.state_in[0]:rnn_state[0],\n",
    "                            self.local_AC.state_in[1]:rnn_state[1]})[0,0]\n",
    "                        v_l,p_l,e_l,g_n,v_n = self.train(episode_buffer,sess,gamma,v1)\n",
    "                        episode_buffer = []\n",
    "                        sess.run(self.update_local_ops)\n",
    "                    if d == True:\n",
    "                        break\n",
    "                                            \n",
    "                self.episode_rewards.append(episode_reward)\n",
    "                self.episode_lengths.append(episode_step_count)\n",
    "                self.episode_mean_values.append(np.mean(episode_values))\n",
    "                \n",
    "                if len(episode_buffer) != 0:\n",
    "                    v_l,p_l,e_l,g_n,v_n = self.train(episode_buffer,sess,gamma,0.0)\n",
    "                     \n",
    "                if episode_count % 50 == 0 and self.name == 'worker_0':\n",
    "                    saver.save(sess,self.model_path+'/model-'+str(episode_count)+'.cptk')\n",
    "                    print (\"Saved Model\")\n",
    "\n",
    "                mean_reward = np.mean(self.episode_rewards[-5:])\n",
    "                mean_length = np.mean(self.episode_lengths[-5:])\n",
    "                mean_value = np.mean(self.episode_mean_values[-5:])\n",
    "                \n",
    "                summary.value.add(tag='env/shares', simple_value=float(self.env.n_shares)) \n",
    "                summary.value.add(tag='Perf/Act', simple_value=float(a)) \n",
    "                summary.value.add(tag='Perf/Episode_reward', simple_value=float(episode_reward)) \n",
    "                summary.value.add(tag='Perf/Reward', simple_value=float(mean_reward))\n",
    "                summary.value.add(tag='Perf/Length', simple_value=float(mean_length))\n",
    "                summary.value.add(tag='Perf/Value', simple_value=float(mean_value))\n",
    "                summary.value.add(tag='Losses/Value Loss', simple_value=float(v_l))\n",
    "                summary.value.add(tag='Losses/Policy Loss', simple_value=float(p_l))\n",
    "                summary.value.add(tag='Losses/Entropy', simple_value=float(e_l))\n",
    "                summary.value.add(tag='Losses/Grad Norm', simple_value=float(g_n))\n",
    "                summary.value.add(tag='Losses/Var Norm', simple_value=float(v_n))\n",
    "                self.summary_writer.add_summary(summary, episode_count)\n",
    "                self.summary_writer.flush()\n",
    "                if self.name == 'worker_0':\n",
    "                    sess.run(self.increment)\n",
    "                episode_count += 1\n",
    "                print(episode_count, episode_reward)\n",
    "\n",
    "                \n",
    "class Test_Worker():\n",
    "    def __init__(self, env, name,s_size,a_size,trainer,model_path,global_episodes):\n",
    "        self.name = \"worker_\" + str(name)\n",
    "        self.number = name        \n",
    "        self.model_path = model_path\n",
    "        self.trainer = trainer\n",
    "        self.global_episodes = global_episodes\n",
    "        self.increment = self.global_episodes.assign_add(1)\n",
    "        self.acts = []\n",
    "        self.rewards = []\n",
    "        self.summary_writer = tf.summary.FileWriter('tb/train_'+str(self.number))\n",
    "        self.local_AC = AC_Network(s_size,a_size,self.name,trainer)\n",
    "        self.update_local_ops = update_target_graph('global',self.name)        \n",
    "        self.env = env\n",
    "        self.actions = [-1, 0, 1]\n",
    "        self.prev_act = 1\n",
    "        \n",
    "    @profile(immediate=True)      \n",
    "    def work(self,max_episode_length,gamma,sess,saver, dep, tresh):\n",
    "        episode_count = sess.run(self.global_episodes)\n",
    "        total_steps = 0\n",
    "        print (\"Starting worker \" + str(self.number))\n",
    "        with sess.as_default(), sess.graph.as_default():    \n",
    "            sess.run(self.update_local_ops)\n",
    "            d = False\n",
    "            episode_reward = 0\n",
    "            s0 = self.env.reset()\n",
    "            action_buffer = [0]*dep\n",
    "            s = np.concatenate((s0,action_buffer))\n",
    "            rnn_state = self.local_AC.state_init\n",
    "            summary = tf.Summary()\n",
    "            \n",
    "            while d == False:\n",
    "                a_dist,v,rnn_state = sess.run([self.local_AC.policy,self.local_AC.value,self.local_AC.state_out], \n",
    "                    feed_dict={self.local_AC.inputs:[s],\n",
    "                    self.local_AC.state_in[0]:rnn_state[0],\n",
    "                    self.local_AC.state_in[1]:rnn_state[1]})\n",
    "                # print(\"A\",a_dist[0], a_dist)\n",
    "                p = np.amax(a_dist[0])\n",
    "                a = np.argmax(a_dist[0])\n",
    "                # print('argmax',p,'max',a)\n",
    "                if(p < tresh):\n",
    "                    a = self.prev_act\n",
    "                self.prev_act = a\n",
    "                # print(a)\n",
    "                # a = np.random.choice(a_dist[0],p=a_dist[0])\n",
    "                # print(\"RC\",a)\n",
    "                # a = np.argmax(a_dist == a)\n",
    "                # print(\"ARGMAX\",a)\n",
    "                s1, r, d, _ = self.env.step(self.actions[a])\n",
    "                action_buffer[:] = np.concatenate((action_buffer[1:],[self.actions[a]]))\n",
    "                s1 = np.concatenate((s1,action_buffer))\n",
    "                if d == True:\n",
    "                    s1 = s\n",
    "                total_steps += 1\n",
    "                episode_reward += r\n",
    "                s = s1\n",
    "                self.acts.append(self.actions[a])\n",
    "                self.rewards.append(r)\n",
    "\n",
    "            self.episode_reward = np.cumsum(self.rewards)\n",
    "            # for i in range(len(self.actions)):\n",
    "            #     summary.value.add(tag='test/a', simple_value=float(self.actions[i])) \n",
    "            #     summary.value.add(tag='test/r', simple_value=float(self.rewards[i])) \n",
    "            #     summary.value.add(tag='test/equity', simple_value=float(self.episode_reward[i])) \n",
    "            #     self.summary_writer.add_summary(summary, i)\n",
    "            #     self.summary_writer.flush()\n",
    "            if self.name == 'worker_0':\n",
    "                sess.run(self.increment)\n",
    "            episode_count += 1\n",
    "            print(total_steps, episode_reward)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-29T16:14:20.556100",
     "start_time": "2017-05-29T16:14:20.240769Z"
    },
    "code_folding": [],
    "collapsed": true,
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "# data\n",
    "postfix = ['RTS-3.16', 'RTS-6.16']\n",
    "R = pd.read_pickle('data/'+str(postfix))\n",
    "test_R = R.copy()\n",
    "# print (R.mean()[0], R.max()[0], (R.max()[0] - R.min()[0]), (R.max()[0] - R.mean()[0])/(R.max()[0] - R.min()[0]), (R.min()[0] - R.mean()[0])/(R.max()[0] - R.min()[0]))\n",
    "R = (R - R.mean()) / (R.max() - R.min())\n",
    "old_R = R.copy()\n",
    "vals = R.values\n",
    "D = np.hstack([vals[i:-dep+i-1, :] for i in range(dep,0,-1)])\n",
    "R = pd.DataFrame(D, R[dep:-1].index)\n",
    "# R = R[:10000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-29T16:14:36.117157",
     "start_time": "2017-05-29T16:14:36.085633Z"
    },
    "code_folding": [],
    "collapsed": true,
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "# Main\n",
    "def run_trades(max_episode_length, gamma, s_size, a_size, load_model, model_path, length, env):\n",
    "    tf.reset_default_graph()\n",
    "    with tf.device('/cpu:0'): \n",
    "        global_episodes = tf.Variable(0,dtype=tf.int32,name='global_episodes',trainable=False)\n",
    "        trainer = tf.train.RMSPropOptimizer(learning_rate = 1e-3, decay = 0.99, epsilon = 1e-6)\n",
    "        master_network = AC_Network(s_size,a_size,'global',None)\n",
    "        num_workers = 1\n",
    "        workers = []\n",
    "        for i in range(num_workers):\n",
    "            workers.append(Test_Worker(env,i,s_size,a_size,trainer,model_path,global_episodes))\n",
    "        saver = tf.train.Saver(max_to_keep=5)\n",
    "\n",
    "    gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0)\n",
    "    if not os.path.exists(model_path):\n",
    "        os.makedirs(model_path)\n",
    "    with tf.Session(config=tf.ConfigProto(allow_soft_placement=False, log_device_placement=False, gpu_options=gpu_options)) as sess:   \n",
    "        if load_model == True:\n",
    "            print ('Loading Model...')\n",
    "            ckpt = tf.train.get_checkpoint_state(model_path)\n",
    "            saver.restore(sess,ckpt.model_checkpoint_path)\n",
    "        else:\n",
    "            sess.run(tf.global_variables_initializer())\n",
    "\n",
    "        summary_writer = tf.summary.FileWriter('tb/train', graph=sess.graph)\n",
    "\n",
    "        workers[0].work(max_episode_length,gamma,sess,saver,dep, 0.33)\n",
    "        acts = workers[0].acts\n",
    "        rews = workers[0].rewards\n",
    "    return acts, rews"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-29T16:14:38.083693",
     "start_time": "2017-05-29T16:14:38.065391Z"
    },
    "code_folding": [],
    "collapsed": true,
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "## Торговая система - покупаем/продаем в зависимости от предсказанного класса\n",
    "def backtest(acts, o_R):\n",
    "    pacts = np.asarray(acts).copy()\n",
    "    for i in range(1,pacts.shape[0]):\n",
    "        if pacts[i] == 0:\n",
    "            if pacts[i-1] == -1:\n",
    "                #from short\n",
    "                pacts[i] = 0.5\n",
    "            elif pacts[i-1] == 1:\n",
    "                pacts[i] = -0.5\n",
    "\n",
    "    # plt.plot(pacts, '.')\n",
    "    data = o_R[dep-1:-3]\n",
    "\n",
    "    buy= cover = pd.Series(pacts == 1, index=data.index)\n",
    "    short = sell= pd.Series(pacts == -1, index=data.index)\n",
    "    cover = pd.Series(pacts > 0, index=data.index)\n",
    "    sell = pd.Series(pacts < 0, index=data.index)\n",
    "    # print(np.sum(buy-sell))\n",
    "    OHLC = data[[('DealPrice', 'close'), ('DealPrice', 'close'), ('DealPrice', 'close'), ('DealPrice', 'close')]]\n",
    "    OHLC.columns = ['O', 'H', 'L', 'C']\n",
    "    # Основной график эквити\n",
    "    bt = pb.Backtest(locals())\n",
    "    return bt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-29T16:14:40.286782",
     "start_time": "2017-05-29T16:14:40.277087Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "load_model = True\n",
    "model_path = 'model'\n",
    "length = R.shape[0]\n",
    "noise_level = 1e-5\n",
    "price = R.values[:,3]\n",
    "n_noises = 1\n",
    "n_runs = 4\n",
    "\n",
    "test_acts = np.zeros((n_noises, n_runs, length-1))\n",
    "test_rews = np.zeros((n_noises, n_runs, length-1))\n",
    "\n",
    "env = environment(R, length)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-29T16:16:24.953994",
     "start_time": "2017-05-29T16:14:41.421737Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading Model...\n",
      "INFO:tensorflow:Restoring parameters from model/model-700.cptk\n",
      "Starting worker 0\n",
      "92582 -10.9924729564\n",
      "\n",
      "*** PROFILER RESULTS ***\n",
      "work (<ipython-input-23-3e21a0f1eb81>:268)\n",
      "function called 1 times\n",
      "\n",
      "         57309032 function calls (56568370 primitive calls) in 97.906 seconds\n",
      "\n",
      "   Ordered by: cumulative time, internal time, call count\n",
      "   List reduced from 132 to 40 due to restriction <40>\n",
      "\n",
      "   ncalls  tottime  percall  cumtime  percall filename:lineno(function)\n",
      "        1    4.074    4.074   97.982   97.982 <ipython-input-23-3e21a0f1eb81>:268(work)\n",
      "    92585    0.774    0.000   88.684    0.001 session.py:675(run)\n",
      "    92585    4.656    0.000   86.858    0.001 session.py:911(_run)\n",
      "    92585    0.161    0.000   39.474    0.000 session.py:1087(_do_run)\n",
      "    92585    0.112    0.000   39.313    0.000 session.py:1134(_do_call)\n",
      "    92585    0.835    0.000   39.201    0.000 session.py:1111(_run_fn)\n",
      "    92585   36.206    0.000   36.206    0.000 {built-in method _pywrap_tensorflow_internal.TF_Run}\n",
      "    92585    2.949    0.000   19.662    0.000 session.py:397(__init__)\n",
      "555501/92585    1.515    0.000   10.948    0.000 session.py:211(for_fetch)\n",
      "185165/92583    0.536    0.000   10.751    0.000 session.py:330(__init__)\n",
      "185165/92583    0.404    0.000    8.911    0.000 session.py:337(<listcomp>)\n",
      "   277746    1.251    0.000    6.816    0.000 tensor_shape.py:675(is_compatible_with)\n",
      "   648082    1.353    0.000    5.335    0.000 ops.py:2381(as_graph_element)\n",
      "   370336    1.025    0.000    5.232    0.000 session.py:247(__init__)\n",
      "   277746    0.466    0.000    4.058    0.000 tensor_shape.py:795(as_shape)\n",
      "   648082    1.425    0.000    3.983    0.000 ops.py:2416(_as_graph_element_locked)\n",
      "   277746    0.807    0.000    3.504    0.000 tensor_shape.py:402(__init__)\n",
      "  9906370    2.440    0.000    3.395    0.000 {built-in method builtins.isinstance}\n",
      "   648082    1.725    0.000    3.382    0.000 ops.py:314(name)\n",
      "    92585    1.068    0.000    3.095    0.000 session.py:455(build_results)\n",
      "    92582    0.769    0.000    2.685    0.000 nest.py:147(flatten_dict_items)\n",
      "   185165    1.475    0.000    2.685    0.000 session.py:291(_uniquify_fetches)\n",
      "   277746    0.398    0.000    2.342    0.000 tensor_shape.py:436(<listcomp>)\n",
      "  1110984    0.869    0.000    2.228    0.000 tensor_shape.py:362(as_dimension)\n",
      "  1573910    0.931    0.000    1.986    0.000 ops.py:1278(name)\n",
      "185165/92583    1.213    0.000    1.787    0.000 session.py:343(build_results)\n",
      "   277747    0.440    0.000    1.681    0.000 numeric.py:463(asarray)\n",
      "    92582    0.467    0.000    1.650    0.000 fromnumeric.py:2154(amax)\n",
      "   185165    1.640    0.000    1.640    0.000 {built-in method numpy.core.multiarray.concatenate}\n",
      "   648082    0.415    0.000    1.637    0.000 ops.py:127(_as_graph_element)\n",
      "    92585    0.504    0.000    1.516    0.000 session.py:1190(_update_with_movers)\n",
      "   185172    0.707    0.000    1.513    0.000 contextlib.py:63(__exit__)\n",
      "  2314582    1.098    0.000    1.463    0.000 python_message.py:647(getter)\n",
      "   277746    0.168    0.000    1.423    0.000 nest.py:74(is_sequence)\n",
      "   925838    1.354    0.000    1.354    0.000 {built-in method builtins.getattr}\n",
      "   370344    0.317    0.000    1.321    0.000 {built-in method builtins.next}\n",
      "   648082    0.657    0.000    1.303    0.000 compat.py:46(as_bytes)\n",
      "   370336    0.307    0.000    1.299    0.000 ops.py:3821(get_default_graph)\n",
      "   277747    1.241    0.000    1.241    0.000 {built-in method numpy.core.multiarray.array}\n",
      "    92582    0.092    0.000    1.182    0.000 _methods.py:25(_amax)\n",
      "\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "([0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  0,\n",
       "  1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  -1,\n",
       "  -1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  -1,\n",
       "  -1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  1,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  -1,\n",
       "  -1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  -1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  -1,\n",
       "  -1,\n",
       "  0,\n",
       "  1,\n",
       "  -1,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  0,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  -1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  1,\n",
       "  ...],\n",
       " [0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  -7.9999999999999996e-06,\n",
       "  -9.0000000000000002e-06,\n",
       "  -1.0000000000000001e-05,\n",
       "  -0.0012978617326448616,\n",
       "  -0.0034423407917383853,\n",
       "  0.0040150642570281067,\n",
       "  0.0040140642570281066,\n",
       "  -0.00042927653471027928,\n",
       "  0.00086058519793461019,\n",
       "  -0.00028786173264486084,\n",
       "  -0.0010020000000000001,\n",
       "  0.00072117039586919087,\n",
       "  -0.00042627653471027926,\n",
       "  -0.00086058519793458244,\n",
       "  0.0011464469305794711,\n",
       "  0.0020060321285140536,\n",
       "  -0.0018635851979345833,\n",
       "  -0.00042627653471025151,\n",
       "  -0.0034423407917383853,\n",
       "  -0.00028786173264488859,\n",
       "  -0.0010020000000000001,\n",
       "  0.00043430866322433006,\n",
       "  0.00057372346528974938,\n",
       "  -0.0017221703958691925,\n",
       "  -0.0022968938611589418,\n",
       "  0.0014313086632243319,\n",
       "  0.0008565851979346102,\n",
       "  -0.00057872346528974939,\n",
       "  -0.00086658519793458247,\n",
       "  -0.0020150321285140535,\n",
       "  0.00085258519793458246,\n",
       "  -0.0014433086632243317,\n",
       "  0.0017111703958691926,\n",
       "  -0.001011,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  -7.9999999999999996e-06,\n",
       "  -9.0000000000000002e-06,\n",
       "  -1.0000000000000001e-05,\n",
       "  -1.1e-05,\n",
       "  -1.2e-05,\n",
       "  -0.00072613826735514093,\n",
       "  -0.001147446930579471,\n",
       "  -0.00057472346528972165,\n",
       "  0.0017191703958691926,\n",
       "  0.001144446930579471,\n",
       "  0.0028646173264486358,\n",
       "  0.00085558519793461018,\n",
       "  -0.0010059999999999999,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -0.0015767234652897233,\n",
       "  0.001147446930579471,\n",
       "  -0.00028786173264488859,\n",
       "  0.00028486173264488857,\n",
       "  0.00085758519793458247,\n",
       "  -0.0011514469305794711,\n",
       "  -0.00029186173264486082,\n",
       "  -0.003735202524383246,\n",
       "  -0.00058072346528972158,\n",
       "  0.0014263086632243318,\n",
       "  -0.0010089999999999999,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  0.00071517039586919083,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -0.0011494469305794711,\n",
       "  -0.0008635851979345824,\n",
       "  -0.00029086173264488856,\n",
       "  0.0028636173264486638,\n",
       "  -0.0010059999999999999,\n",
       "  0.0,\n",
       "  0.0012938938611589126,\n",
       "  0.0057372346528972995,\n",
       "  0.0014333086632243319,\n",
       "  0.0017191703958691926,\n",
       "  -0.001003,\n",
       "  0.0,\n",
       "  -0.0021484469305794729,\n",
       "  -0.0017211703958691926,\n",
       "  -9.9999999999999995e-07,\n",
       "  0.00057172346528974933,\n",
       "  0.0028656173264486359,\n",
       "  -0.0010040000000000001,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -0.0010020000000000018,\n",
       "  0.0034423407917383853,\n",
       "  -0.0017221703958691925,\n",
       "  0.00085858519793461014,\n",
       "  -0.001003,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  0.001577755593803801,\n",
       "  -0.001147446930579471,\n",
       "  0.0011464469305794711,\n",
       "  0.0031534790590935246,\n",
       "  0.0014313086632243041,\n",
       "  -0.0025857555938037752,\n",
       "  0.001142446930579471,\n",
       "  -0.00086658519793461023,\n",
       "  -6.9999999999999999e-06,\n",
       "  -0.0014423086632243318,\n",
       "  -0.00086958519793458244,\n",
       "  -0.00101,\n",
       "  0.0,\n",
       "  0.00043330866322433003,\n",
       "  -0.00057372346528974938,\n",
       "  -0.0022958938611589144,\n",
       "  -0.00086258519793461024,\n",
       "  0.001144446930579471,\n",
       "  -0.0010040000000000001,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  -7.9999999999999996e-06,\n",
       "  0.0024333407917383558,\n",
       "  -0.0014343086632243041,\n",
       "  -0.0005747234652897494,\n",
       "  0.00057172346528974933,\n",
       "  -3.0000000000000001e-06,\n",
       "  -0.00057772346528974937,\n",
       "  -0.002299893861158914,\n",
       "  0.0014283086632243319,\n",
       "  0.0002798617326448608,\n",
       "  -0.001008,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -0.0010050000000000018,\n",
       "  -0.0014343086632243318,\n",
       "  0.0005727234652897216,\n",
       "  -0.00028886173264486081,\n",
       "  0.00028386173264486079,\n",
       "  -0.0014383086632243319,\n",
       "  -0.00057872346528972164,\n",
       "  -0.0010059999999999999,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  0.00014344693057946926,\n",
       "  0.00028686173264486081,\n",
       "  -0.00086158519793461022,\n",
       "  -0.0017231703958691927,\n",
       "  0.00028386173264488855,\n",
       "  0.001430308663224304,\n",
       "  -0.002299893861158914,\n",
       "  -0.0017271703958691926,\n",
       "  0.0034353407917383852,\n",
       "  0.00056572346528972165,\n",
       "  0.0031464790590935245,\n",
       "  -0.00101,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -0.0015777234652897235,\n",
       "  0.00057372346528972162,\n",
       "  0.0014333086632243319,\n",
       "  -0.00028886173264486081,\n",
       "  0.00085758519793458247,\n",
       "  -0.0002908617326448608,\n",
       "  0.0017161703958691926,\n",
       "  -0.00057972346528972166,\n",
       "  -0.0010070000000000001,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -0.00014341480206539155,\n",
       "  0.00028686173264486081,\n",
       "  0.0022938938611589141,\n",
       "  0.00028486173264486082,\n",
       "  0.0017181703958691927,\n",
       "  0.0017171703958691925,\n",
       "  -0.0031604790590934965,\n",
       "  0.00028086173264486083,\n",
       "  -0.0010070000000000001,\n",
       "  0.0,\n",
       "  0.00072017039586919085,\n",
       "  -0.0020080321285140534,\n",
       "  -0.00028786173264486084,\n",
       "  -0.0011494469305794711,\n",
       "  -0.00028986173264486083,\n",
       "  -0.0017251703958691927,\n",
       "  -0.0017261703958691926,\n",
       "  0.00056772346528974934,\n",
       "  0.0020010321285140534,\n",
       "  -0.001008,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -0.0015767234652897233,\n",
       "  0.0020080321285140534,\n",
       "  -0.00086158519793461022,\n",
       "  0.00057172346528974933,\n",
       "  0.00085758519793458247,\n",
       "  -0.0011514469305794433,\n",
       "  -0.0031604790590935242,\n",
       "  -0.0023008938611589142,\n",
       "  -0.0034493407917383853,\n",
       "  0.0011394469305794433,\n",
       "  0.00056472346528974938,\n",
       "  0.0014243086632243318,\n",
       "  0.00056272346528972157,\n",
       "  -0.0017331703958691927,\n",
       "  -0.0020210321285140534,\n",
       "  0.00055972346528972161,\n",
       "  0.0031404790590935246,\n",
       "  -1.5999999999999999e-05,\n",
       "  0.0017041703958691925,\n",
       "  -0.0010180000000000009,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -0.0018635851979345833,\n",
       "  -0.00028686173264486081,\n",
       "  0.0014333086632243041,\n",
       "  -0.0011494469305794433,\n",
       "  -0.0014373086632243318,\n",
       "  0.00056972346528972163,\n",
       "  -0.002299893861158914,\n",
       "  0.0014283086632243319,\n",
       "  -0.00058072346528974933,\n",
       "  0.0005657234652897494,\n",
       "  -9.0000000000000002e-06,\n",
       "  -0.0028786173264486637,\n",
       "  0.0014233086632243319,\n",
       "  -0.0014463086632243319,\n",
       "  -1.2999999999999999e-05,\n",
       "  -0.00087458519793458245,\n",
       "  0.0008455851979345824,\n",
       "  0.00055772346528974942,\n",
       "  -0.0011644469305794711,\n",
       "  0.00026886173264486081,\n",
       "  0.00026786173264486078,\n",
       "  0.0017011703958691926,\n",
       "  0.0014133086632243318,\n",
       "  -0.0003088617326448608,\n",
       "  0.00026386173264486079,\n",
       "  -0.0003108617326448608,\n",
       "  -0.00031186173264486082,\n",
       "  0.0028426173264486359,\n",
       "  0.00083358519793461019,\n",
       "  0.0011194469305794432,\n",
       "  -0.00060272346528972168,\n",
       "  -0.0014643086632243319,\n",
       "  0.0014033086632243318,\n",
       "  0.0048446494549627172,\n",
       "  -0.00060672346528974942,\n",
       "  -0.0020420321285140536,\n",
       "  -0.0003218617326448608,\n",
       "  0.00082458519793461018,\n",
       "  -0.00061072346528974941,\n",
       "  -0.0011854469305794433,\n",
       "  -0.0017601703958691926,\n",
       "  -4.0000000000000003e-05,\n",
       "  -0.0006147234652897494,\n",
       "  0.00053172346528974933,\n",
       "  -0.00032986173264486083,\n",
       "  -0.00061772346528974936,\n",
       "  0.00081558519793461018,\n",
       "  0.0002408617326448608,\n",
       "  -4.7000000000000004e-05,\n",
       "  -0.001048000000000001,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  -7.9999999999999996e-06,\n",
       "  -9.0000000000000002e-06,\n",
       "  -1.0000000000000001e-05,\n",
       "  -1.1e-05,\n",
       "  -1.2e-05,\n",
       "  -1.2999999999999999e-05,\n",
       "  -1.4e-05,\n",
       "  -1.5e-05,\n",
       "  -1.5999999999999999e-05,\n",
       "  -1.7000000000000003e-05,\n",
       "  -1.8e-05,\n",
       "  -1.9000000000000001e-05,\n",
       "  -2.0000000000000002e-05,\n",
       "  -2.1000000000000002e-05,\n",
       "  -2.1999999999999999e-05,\n",
       "  -2.3e-05,\n",
       "  -2.4000000000000001e-05,\n",
       "  -2.5000000000000001e-05,\n",
       "  -2.5999999999999998e-05,\n",
       "  -2.6999999999999999e-05,\n",
       "  -2.8e-05,\n",
       "  -2.9000000000000004e-05,\n",
       "  -3.0000000000000001e-05,\n",
       "  -3.1000000000000001e-05,\n",
       "  -3.1999999999999999e-05,\n",
       "  -3.3000000000000003e-05,\n",
       "  -3.4000000000000007e-05,\n",
       "  -3.5000000000000004e-05,\n",
       "  -3.6000000000000001e-05,\n",
       "  -3.6999999999999998e-05,\n",
       "  -3.8000000000000002e-05,\n",
       "  -3.8999999999999999e-05,\n",
       "  -4.0000000000000003e-05,\n",
       "  -4.1e-05,\n",
       "  -4.2000000000000004e-05,\n",
       "  -4.2999999999999995e-05,\n",
       "  -0.0013308617326448617,\n",
       "  0.0017211703958691926,\n",
       "  -0.00014041480206539069,\n",
       "  -0.00057372346528974938,\n",
       "  -0.00086158519793458246,\n",
       "  -0.0010020000000000009,\n",
       "  0.0,\n",
       "  -0.001861585197934611,\n",
       "  0.00057372346528972162,\n",
       "  0.00028586173264488854,\n",
       "  -0.00086258519793461024,\n",
       "  -0.0014373086632243318,\n",
       "  0.0008565851979346102,\n",
       "  -0.00057872346528974939,\n",
       "  -0.0010060000000000008,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -0.0012888617326448617,\n",
       "  0.00086058519793458244,\n",
       "  0.0011464469305794711,\n",
       "  -0.0010020000000000009,\n",
       "  0.0,\n",
       "  0.0005727234652897216,\n",
       "  -0.0010020000000000009,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -0.00014141480206541845,\n",
       "  0.0,\n",
       "  -0.00057472346528972165,\n",
       "  -0.00057572346528972167,\n",
       "  0.0014313086632243319,\n",
       "  -0.00057772346528974937,\n",
       "  0.002576755593803803,\n",
       "  -0.0010060000000000008,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -0.0015757234652897226,\n",
       "  0.0025817555938037751,\n",
       "  -0.0010010000000000008,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  -7.9999999999999996e-06,\n",
       "  -9.0000000000000002e-06,\n",
       "  -0.00014941480206541845,\n",
       "  0.00086058519793461019,\n",
       "  -0.0012878617326448616,\n",
       "  -0.0017211703958691926,\n",
       "  -9.9999999999999995e-07,\n",
       "  0.00028486173264486082,\n",
       "  -3.0000000000000001e-06,\n",
       "  -0.003298893861158943,\n",
       "  -0.0045897877223178285,\n",
       "  0.0025807555938037749,\n",
       "  -0.0024363086632243325,\n",
       "  -0.0010000000000000009,\n",
       "  -0.0012868617326448617,\n",
       "  0.00086058519793461019,\n",
       "  0.0014333086632243041,\n",
       "  0.00057172346528974933,\n",
       "  -0.0010030000000000009,\n",
       "  0.0,\n",
       "  -0.0044433407917383863,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -0.0017231703958691927,\n",
       "  0.0020050321285140535,\n",
       "  -0.002872617326448636,\n",
       "  0.0011424469305794432,\n",
       "  -0.0010060000000000008,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -0.0010020000000000009,\n",
       "  0.00028686173264486081,\n",
       "  0.0020070321285140533,\n",
       "  -1.9999999999999999e-06,\n",
       "  0.0020050321285140535,\n",
       "  0.00028286173264486082,\n",
       "  0.00028186173264488855,\n",
       "  0.0008545851979345824,\n",
       "  -0.00058072346528972158,\n",
       "  0.0031474790590934965,\n",
       "  -0.0017301703958691648,\n",
       "  0.001424308663224304,\n",
       "  0.0028576173264486635,\n",
       "  -0.00058572346528973547,\n",
       "  -0.0008735851979345963,\n",
       "  0.0002728617326448608,\n",
       "  -0.0011624469305794433,\n",
       "  0.0019920321285140396,\n",
       "  -0.00087758519793459629,\n",
       "  -0.002312893861158914,\n",
       "  0.0005547234652897216,\n",
       "  -0.0023148938611589143,\n",
       "  -0.001168446930579471,\n",
       "  0.0014123086632243317,\n",
       "  -0.00088358519793458245,\n",
       "  -0.00059772346528974942,\n",
       "  -0.00031186173264486082,\n",
       "  -0.0010260000000000009,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -0.00014641480206540458,\n",
       "  0.001147446930579471,\n",
       "  0.0022938938611589141,\n",
       "  -0.00028886173264486081,\n",
       "  0.002578755593803789,\n",
       "  0.002577755593803775,\n",
       "  -0.0028736173264486495,\n",
       "  0.00056772346528973546,\n",
       "  -0.002588755593803789,\n",
       "  0.0020000321285140532,\n",
       "  0.002572755593803789,\n",
       "  -0.00058372346528973553,\n",
       "  0.0028576173264486635,\n",
       "  -0.0025937555938037888,\n",
       "  0.0011344469305794571,\n",
       "  -0.00058772346528973552,\n",
       "  -0.0020230321285140533,\n",
       "  -0.00058972346528972158,\n",
       "  -0.00374620252438326,\n",
       "  -1.8e-05,\n",
       "  0.00055472346528973547,\n",
       "  0.0034223407917383852,\n",
       "  -0.00059472346528973547,\n",
       "  -0.0046117877223178427,\n",
       "  -0.00030986173264486083,\n",
       "  0.00054972346528973546,\n",
       "  -2.5000000000000001e-05,\n",
       "  -0.00059972346528973548,\n",
       "  -0.0017481703958691925,\n",
       "  0.0022668938611589283,\n",
       "  -0.00088958519793459637,\n",
       "  -0.001030000000000001,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  -7.9999999999999996e-06,\n",
       "  -9.0000000000000002e-06,\n",
       "  -1.0000000000000001e-05,\n",
       "  -1.1e-05,\n",
       "  -0.0021594469305794579,\n",
       "  -0.00086058519793461019,\n",
       "  0.0011464469305794711,\n",
       "  0.0014323086632243318,\n",
       "  0.0025787555938037751,\n",
       "  0.00056972346528973551,\n",
       "  -0.00071813826735515397,\n",
       "  0.0011474469305794571,\n",
       "  -0.00014041480206541845,\n",
       "  -0.00086058519793458244,\n",
       "  -0.00057472346528973553,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -0.0008645851979345963,\n",
       "  -0.00057872346528973551,\n",
       "  0.00028086173264487471,\n",
       "  -6.9999999999999999e-06,\n",
       "  0.0011394469305794571,\n",
       "  0.00027786173264486081,\n",
       "  -0.00058372346528972165,\n",
       "  -0.0010110000000000008,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  -7.9999999999999996e-06,\n",
       "  -9.0000000000000002e-06,\n",
       "  -1.0000000000000001e-05,\n",
       "  -1.1e-05,\n",
       "  -1.2e-05,\n",
       "  -1.2999999999999999e-05,\n",
       "  -1.4e-05,\n",
       "  -1.5e-05,\n",
       "  -1.5999999999999999e-05,\n",
       "  -1.7000000000000003e-05,\n",
       "  -1.8e-05,\n",
       "  -1.9000000000000001e-05,\n",
       "  -2.0000000000000002e-05,\n",
       "  -2.1000000000000002e-05,\n",
       "  -2.1999999999999999e-05,\n",
       "  -2.3e-05,\n",
       "  -2.4000000000000001e-05,\n",
       "  -2.5000000000000001e-05,\n",
       "  -2.5999999999999998e-05,\n",
       "  -2.6999999999999999e-05,\n",
       "  -2.8e-05,\n",
       "  -2.9000000000000004e-05,\n",
       "  -3.0000000000000001e-05,\n",
       "  -3.1000000000000001e-05,\n",
       "  -3.1999999999999999e-05,\n",
       "  -3.3000000000000003e-05,\n",
       "  -3.4000000000000007e-05,\n",
       "  -3.5000000000000004e-05,\n",
       "  -3.6000000000000001e-05,\n",
       "  -3.6999999999999998e-05,\n",
       "  -3.8000000000000002e-05,\n",
       "  -3.8999999999999999e-05,\n",
       "  -4.0000000000000003e-05,\n",
       "  -0.00018041480206540457,\n",
       "  0.00086058519793459631,\n",
       "  0.00057272346528973548,\n",
       "  0.00085858519793459626,\n",
       "  -0.0014373086632243318,\n",
       "  -0.001438308663224318,\n",
       "  -0.00057872346528973551,\n",
       "  0.0014283086632243319,\n",
       "  -0.00058072346528973545,\n",
       "  0.0011394469305794571,\n",
       "  -0.00058272346528972163,\n",
       "  0.0019980321285140534,\n",
       "  -0.00029786173264487469,\n",
       "  -1.2e-05,\n",
       "  -0.0002998617326448608,\n",
       "  -0.00058772346528973552,\n",
       "  -1.5e-05,\n",
       "  -1.5999999999999999e-05,\n",
       "  0.00026986173264487471,\n",
       "  0.0011294469305794571,\n",
       "  -0.0014533086632243317,\n",
       "  0.00084058519793459626,\n",
       "  0.00055272346528973553,\n",
       "  -2.1999999999999999e-05,\n",
       "  -0.0011704469305794571,\n",
       "  0.0025577555938037888,\n",
       "  0.0025567555938037752,\n",
       "  0.00054772346528973552,\n",
       "  -0.00060072346528973551,\n",
       "  -0.00088858519793458247,\n",
       "  0.00025786173264486081,\n",
       "  -0.00089058519793459628,\n",
       "  -0.0011784469305794571,\n",
       "  -0.00031886173264487471,\n",
       "  -0.001467308663224318,\n",
       "  0.00053972346528972166,\n",
       "  0.00025186173264487471,\n",
       "  0.0019720321285140534,\n",
       "  0.0011104469305794572,\n",
       "  -0.00061172346528972168,\n",
       "  0.0022558938611589143,\n",
       "  -0.00032686173264486081,\n",
       "  -0.00061472346528973552,\n",
       "  0.0022528938611589282,\n",
       "  -0.00061672346528973546,\n",
       "  0.0011034469305794571,\n",
       "  0.0022498938611589282,\n",
       "  -0.0010460000000000009,\n",
       "  0.0,\n",
       "  -0.0015747234652897224,\n",
       "  -0.0010000000000000009,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  -7.9999999999999996e-06,\n",
       "  -9.0000000000000002e-06,\n",
       "  -1.0000000000000001e-05,\n",
       "  -1.1e-05,\n",
       "  -1.2e-05,\n",
       "  -1.2999999999999999e-05,\n",
       "  -1.4e-05,\n",
       "  -1.5e-05,\n",
       "  -1.5999999999999999e-05,\n",
       "  -1.7000000000000003e-05,\n",
       "  0.00012944693057945623,\n",
       "  0.0,\n",
       "  -0.00028786173264486084,\n",
       "  0.0014323086632243179,\n",
       "  -0.00086358519793459628,\n",
       "  -0.00057772346528972161,\n",
       "  0.0020030321285140536,\n",
       "  -0.0017271703958691926,\n",
       "  0.00085358519793459636,\n",
       "  -0.00029486173264486079,\n",
       "  0.0022858938611589144,\n",
       "  0.0011374469305794571,\n",
       "  -0.00029786173264486081,\n",
       "  0.0011354469305794571,\n",
       "  -0.00058672346528972161,\n",
       "  -0.0017351703958691925,\n",
       "  -0.00087558519793459635,\n",
       "  0.0002708617326448608,\n",
       "  -0.002598755593803789,\n",
       "  0.00285061732644865,\n",
       "  0.00055472346528973547,\n",
       "  -0.00030686173264486081,\n",
       "  -0.0023158938611589283,\n",
       "  -0.0020300321285140533,\n",
       "  0.0016981703958691926,\n",
       "  0.00026286173264486082,\n",
       "  0.00054872346528973554,\n",
       "  0.00083458519793459633,\n",
       "  -2.6999999999999999e-05,\n",
       "  -0.0011754469305794572,\n",
       "  0.0019790321285140535,\n",
       "  -0.0014643086632243319,\n",
       "  0.0019770321285140532,\n",
       "  -0.0014663086632243178,\n",
       "  0.00082758519793459627,\n",
       "  0.0002528617326448608,\n",
       "  0.001112446930579457,\n",
       "  0.00053772346528973549,\n",
       "  -0.001184446930579457,\n",
       "  -3.8000000000000002e-05,\n",
       "  0.0019690321285140535,\n",
       "  0.0005337234652897355,\n",
       "  -0.00061472346528973552,\n",
       "  -0.0017631703958691926,\n",
       "  -0.00061672346528973546,\n",
       "  0.0016771703958691926,\n",
       "  -4.4999999999999996e-05,\n",
       "  -0.00090658519793459635,\n",
       "  0.00023986173264487469,\n",
       "  -0.00090858519793459629,\n",
       "  -0.00033586173264487469,\n",
       "  -5.0000000000000002e-05,\n",
       "  -5.1e-05,\n",
       "  -0.0011994469305794571,\n",
       "  -0.0017741703958691927,\n",
       "  0.0016671703958691926,\n",
       "  0.0010924469305794572,\n",
       "  -0.001490308663224318,\n",
       "  -0.0012044469305794571,\n",
       "  -0.00034486173264487469,\n",
       "  0.00051472346528973548,\n",
       "  0.00051372346528973545,\n",
       "  -6.0999999999999999e-05,\n",
       "  -0.001209446930579471,\n",
       "  0.00051072346528973549,\n",
       "  -0.00092458519793459635,\n",
       "  -0.00063872346528973546,\n",
       "  0.00022086173264487468,\n",
       "  0.00079358519793459631,\n",
       "  -0.00035486173264487472,\n",
       "  -6.900000000000001e-05,\n",
       "  0.00050372346528973553,\n",
       "  0.0013633086632243181,\n",
       "  -7.2000000000000002e-05,\n",
       "  -7.2999999999999999e-05,\n",
       "  -0.00036086173264486082,\n",
       "  -7.4999999999999993e-05,\n",
       "  0.0019320321285140533,\n",
       "  -0.00065072346528972165,\n",
       "  -0.00036486173264487469,\n",
       "  0.0010684469305794709,\n",
       "  -0.00036686173264487468,\n",
       "  -0.00094158519793459633,\n",
       "  0.00049172346528973546,\n",
       "  -0.00094358519793459627,\n",
       "  -0.00065772346528973548,\n",
       "  0.00020186173264487468,\n",
       "  0.00048772346528972165,\n",
       "  -0.00066072346528972168,\n",
       "  -0.00066172346528973547,\n",
       "  -0.00037586173264486081,\n",
       "  -8.9999999999999992e-05,\n",
       "  0.00019586173264486082,\n",
       "  0.00048172346528973549,\n",
       "  0.001341308663224318,\n",
       "  -9.4000000000000008e-05,\n",
       "  0.0027736173264486497,\n",
       "  0.00076458519793459636,\n",
       "  -0.001531308663224318,\n",
       "  -0.00038486173264486081,\n",
       "  0.0004747234652897216,\n",
       "  -0.0012474469305794572,\n",
       "  -0.000101,\n",
       "  0.001045446930579457,\n",
       "  0.0001838617326448608,\n",
       "  0.00018286173264487471,\n",
       "  -0.00039186173264487469,\n",
       "  -0.00096658519793458241,\n",
       "  -0.00096758519793459631,\n",
       "  0.00075258519793459629,\n",
       "  -0.00010900000000000001,\n",
       "  -0.00039686173264487471,\n",
       "  0.00074958519793459633,\n",
       "  0.0001748617326448608,\n",
       "  0.00046072346528973547,\n",
       "  -0.00040086173264486082,\n",
       "  0.00017186173264486081,\n",
       "  0.0004577234652897355,\n",
       "  -0.00097758519793459623,\n",
       "  0.00016886173264486082,\n",
       "  -0.00097958519793459628,\n",
       "  0.00016686173264486082,\n",
       "  -0.00098158519793458245,\n",
       "  0.00045172346528972164,\n",
       "  0.00073758519793459625,\n",
       "  -0.00041086173264486079,\n",
       "  -0.00098558519793459642,\n",
       "  -0.00069972346528973553,\n",
       "  0.0013073086632243319,\n",
       "  -0.00041486173264487466,\n",
       "  -0.0004158617326448608,\n",
       "  -0.00041686173264486083,\n",
       "  0.00044272346528972158,\n",
       "  0.00044172346528973549,\n",
       "  -0.00041986173264486079,\n",
       "  0.00015286173264486081,\n",
       "  -0.000135,\n",
       "  -0.0012834469305794572,\n",
       "  0.00072358519793459634,\n",
       "  -0.00071172346528973549,\n",
       "  0.00014786173264486079,\n",
       "  -0.00014000000000000001,\n",
       "  0.0001458617326448747,\n",
       "  -0.00014199999999999998,\n",
       "  0.00014386173264486083,\n",
       "  0.00014286173264486081,\n",
       "  -0.00043186173264486081,\n",
       "  0.00014086173264486081,\n",
       "  -0.000147,\n",
       "  -0.00014799999999999999,\n",
       "  -0.0011490000000000009,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -0.0018655851979346111,\n",
       "  0.00086058519793459631,\n",
       "  -0.00028786173264486084,\n",
       "  -0.00028886173264486081,\n",
       "  -3.0000000000000001e-06,\n",
       "  -0.00057772346528973549,\n",
       "  -0.00029186173264486082,\n",
       "  0.00085458519793459628,\n",
       "  0.0002798617326448608,\n",
       "  -0.00029486173264486079,\n",
       "  -0.00029586173264487469,\n",
       "  -0.00029686173264486084,\n",
       "  0.00027586173264486082,\n",
       "  -0.00058572346528972159,\n",
       "  0.00027386173264486082,\n",
       "  0.00055972346528973549,\n",
       "  -1.5e-05,\n",
       "  0.00055772346528972167,\n",
       "  -1.7000000000000003e-05,\n",
       "  -0.00030486173264486082,\n",
       "  -0.00087958519793459634,\n",
       "  -0.00059372346528973555,\n",
       "  0.010306022375215156,\n",
       "  -0.0027431703958691934,\n",
       "  0.0010080321285140526,\n",
       "  -0.0011474469305794571,\n",
       "  0.002293893861158928,\n",
       "  0.00028486173264486082,\n",
       "  -0.0020110321285140673,\n",
       "  -0.0011514469305794572,\n",
       "  0.00071617039586919173,\n",
       "  -0.0014343086632243318,\n",
       "  -0.00028786173264486084,\n",
       "  -0.00086258519793459636,\n",
       "  0.00057072346528973554,\n",
       "  0.00028286173264486082,\n",
       "  -0.00029186173264486082,\n",
       "  -0.0025877555938038027,\n",
       "  -0.0020150321285140535,\n",
       "  -0.00086858519793458241,\n",
       "  0.00085158519793458243,\n",
       "  -0.0018705851979346111,\n",
       "  -0.00086058519793458244,\n",
       "  0.0017201703958691927,\n",
       "  -0.0011494469305794711,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  0.001142446930579471,\n",
       "  0.0025757555938037751,\n",
       "  -0.00058072346528972158,\n",
       "  -0.001155446930579471,\n",
       "  0.0019990321285140535,\n",
       "  -0.00087058519793458246,\n",
       "  0.0019970321285140533,\n",
       "  0.0011354469305794709,\n",
       "  0.00084758519793458244,\n",
       "  -0.0020220321285140536,\n",
       "  0.0008455851979345824,\n",
       "  0.00013144693057944235,\n",
       "  -0.00057372346528972162,\n",
       "  0.0005727234652897216,\n",
       "  0.00028486173264486082,\n",
       "  -0.0008635851979345824,\n",
       "  -0.0020120321285140535,\n",
       "  0.0002818617326448608,\n",
       "  0.00028086173264486083,\n",
       "  -6.9999999999999999e-06,\n",
       "  0.0031474790590935242,\n",
       "  -0.0010090000000000008,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  -7.9999999999999996e-06,\n",
       "  -9.0000000000000002e-06,\n",
       "  -1.0000000000000001e-05,\n",
       "  -1.1e-05,\n",
       "  -1.2e-05,\n",
       "  -1.2999999999999999e-05,\n",
       "  -1.4e-05,\n",
       "  -1.5e-05,\n",
       "  -1.5999999999999999e-05,\n",
       "  -1.7000000000000003e-05,\n",
       "  -1.8e-05,\n",
       "  -1.9000000000000001e-05,\n",
       "  -2.0000000000000002e-05,\n",
       "  -2.1000000000000002e-05,\n",
       "  -2.1999999999999999e-05,\n",
       "  -2.3e-05,\n",
       "  -2.4000000000000001e-05,\n",
       "  -0.0033198938611589289,\n",
       "  -0.0025817555938037889,\n",
       "  -0.0022958938611589144,\n",
       "  0.00028486173264486082,\n",
       "  -0.00057672346528973547,\n",
       "  -0.0011514469305794572,\n",
       "  0.0034373407917383855,\n",
       "  -0.0011534469305794571,\n",
       "  0.0017141703958691926,\n",
       "  0.0020000321285140532,\n",
       "  -0.0010090000000000008,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  -7.9999999999999996e-06,\n",
       "  -9.0000000000000002e-06,\n",
       "  -1.0000000000000001e-05,\n",
       "  -1.1e-05,\n",
       "  -1.2e-05,\n",
       "  -1.2999999999999999e-05,\n",
       "  -1.4e-05,\n",
       "  -1.5e-05,\n",
       "  0.00013144693057947011,\n",
       "  -0.0017211703958691926,\n",
       "  -0.0005747234652897494,\n",
       "  0.001145446930579471,\n",
       "  -3.0000000000000001e-06,\n",
       "  -0.00057772346528972161,\n",
       "  0.0020030321285140536,\n",
       "  0.0011414469305794433,\n",
       "  0.00056672346528973555,\n",
       "  -0.0010080000000000009,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -0.0021544469305794581,\n",
       "  -0.00086058519793459631,\n",
       "  0.00057272346528973548,\n",
       "  0.0028666173264486499,\n",
       "  0.00057072346528973554,\n",
       "  -0.0011514469305794711,\n",
       "  -0.0028736173264486495,\n",
       "  -0.0017271703958691926,\n",
       "  0.0011404469305794709,\n",
       "  0.0017131703958691788,\n",
       "  -0.00029586173264486081,\n",
       "  0.00056372346528973548,\n",
       "  0.0011364469305794572,\n",
       "  0.0048646494549627033,\n",
       "  -0.0008735851979345963,\n",
       "  -0.0015877234652897502,\n",
       "  -0.0040160642570281208,\n",
       "  -0.0010010000000000008,\n",
       "  -0.0044423407917383723,\n",
       "  0.0004343086632243448,\n",
       "  -0.0010000000000000009,\n",
       "  0.0,\n",
       "  0.0015807555938037881,\n",
       "  -0.0022948938611589281,\n",
       "  0.0025807555938037888,\n",
       "  0.0014323086632243179,\n",
       "  -0.0014373086632243179,\n",
       "  0.00056972346528972163,\n",
       "  -0.00086558519793458245,\n",
       "  -0.0017271703958691926,\n",
       "  -0.0023018938611589282,\n",
       "  0.0014263086632243318,\n",
       "  0.00056472346528972162,\n",
       "  -0.0017311703958691927,\n",
       "  -1.1e-05,\n",
       "  -1.2e-05,\n",
       "  -0.0008735851979345963,\n",
       "  -0.00058772346528972164,\n",
       "  -0.00087558519793459635,\n",
       "  -0.00030286173264486082,\n",
       "  0.00055672346528972164,\n",
       "  -0.0010180000000000009,\n",
       "  0.0,\n",
       "  -9.9999999999999995e-07,\n",
       "  -1.9999999999999999e-06,\n",
       "  -3.0000000000000001e-06,\n",
       "  -3.9999999999999998e-06,\n",
       "  -5.0000000000000004e-06,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  -7.9999999999999996e-06,\n",
       "  -9.0000000000000002e-06,\n",
       "  0.00042430866322431702,\n",
       "  0.0028686173264486636,\n",
       "  0.0017201703958691927,\n",
       "  0.0034403407917383715,\n",
       "  -0.0017241703958691926,\n",
       "  0.00056972346528973551,\n",
       "  -0.00057872346528973551,\n",
       "  -6.0000000000000002e-06,\n",
       "  -6.9999999999999999e-06,\n",
       "  0.0020000321285140671,\n",
       "  -0.0015827234652897085,\n",
       "  0.0,\n",
       "  0.0025807555938037888,\n",
       "  -0.0020100321285140671,\n",
       "  -0.0021504469305794719,\n",
       "  0.0014343086632243318,\n",
       "  0.00028586173264486079,\n",
       "  0.00057172346528973545,\n",
       "  -0.00086358519793459628,\n",
       "  -0.0002908617326448608,\n",
       "  -0.0002918617326448747,\n",
       "  -0.0014403086632243179,\n",
       "  0.0011404469305794571,\n",
       "  -0.0011554469305794571,\n",
       "  -0.00029586173264487469,\n",
       "  -0.00087058519793459634,\n",
       "  -1.1e-05,\n",
       "  -0.00029886173264486083,\n",
       "  ...])"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "run_trades(max_episode_length, gamma, s_size, a_size, load_model, model_path, length, env)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-07T16:14:49.241622",
     "start_time": "2017-05-07T16:09:23.493822Z"
    },
    "collapsed": true,
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "for j in range(n_noises):\n",
    "    np.random.seed(j)\n",
    "    env = environment(R+(np.random.rand(R.shape[0],R.shape[1])-0.5)*(1e-5*(0**j)), length)\n",
    "    for i in range(n_runs):\n",
    "        np.random.seed(i+1377)\n",
    "        test_acts[j,i,:], test_rews[j,i,:] = run_trades(max_episode_length, gamma, s_size, a_size, load_model, model_path, length, env)\n",
    "        print(j,i)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-07T16:14:50.618585",
     "start_time": "2017-05-07T16:14:49.245300Z"
    },
    "collapsed": true,
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "## ПРИ НЕУВЕРЕННОСТИ НУЖНО ДЕЛАТЬ ПРЕДЫДУЩЕЕ ДЕЙСТВИЕ, А НЕ 0\n",
    "\n",
    "test_acts_probs = np.zeros((n_noises, a_size, length-1))\n",
    "tresh = 0.0\n",
    "acts = np.zeros((n_noises, length-1))\n",
    "test_acts = test_acts.astype(int)\n",
    "for k in range(n_noises):\n",
    "    for j in range(n_runs):\n",
    "        for i in range(length-1):\n",
    "            a = test_acts[k,j,i]\n",
    "            test_acts_probs[k,a+1,i] += 1\n",
    "    test_acts_probs[k,:,:] = test_acts_probs[k,:,:]/n_runs\n",
    "    acts[k,:] = [np.argmax(x)-1 if max(x) > tresh else 0 for x in test_acts_probs[k,:,:].T]\n",
    "# print(test_acts_probs)\n",
    "# plt.plot(acts.T)\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-07T16:26:58.262996",
     "start_time": "2017-05-07T16:26:56.338703Z"
    },
    "collapsed": true,
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "def write_report(r, filename):    \n",
    "    with open(filename, \"a\") as input_file:\n",
    "        for k, v in r.items():\n",
    "            line = '{}, {}'.format(k, v) \n",
    "            print(line, file=input_file)\n",
    "            \n",
    "folder = os.path.relpath(\".\",\"..\")\n",
    "for i in range(4):    \n",
    "    lengths = [ R.index.get_loc('2015-12-29 10:06:00')[0], R.index.get_loc('2016-01-15 10:06:00')[0], \\\n",
    "                    R.index.get_loc('2016-03-15 10:06:00')[0], R.index.get_loc('2016-06-15 10:06:00')[0] ]\n",
    "    names = ['2w', '1m', '3m', '6m']\n",
    "    bt = backtest(acts[-1,:lengths[i]], test_R[:lengths[i]+dep+2])\n",
    "    fig = plt.figure(figsize=(12, 10))\n",
    "    bt.plot_trades()\n",
    "    plt.legend(loc='best')\n",
    "    plt.title('Backtesting'+folder+'_'+names[i])\n",
    "    plt.ylabel('Rubles')\n",
    "    plt.xlabel('Time')        \n",
    "    plt.plot()\n",
    "    # plt.show()\n",
    "    plt.savefig('../plots/'+folder+'_'+names[i]+'.png', bbox_inches='tight')\n",
    "\n",
    "    # Отчет по системе\n",
    "    B = bt.report\n",
    "    write_report(B, '../plots/'+folder+'_.txt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}