techstock.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:20:40.826355Z",
     "start_time": "2020-06-17T15:20:40.435589Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn import preprocessing\n",
    "from sklearn.utils import shuffle\n",
    "import seaborn as sns\n",
    "import pickle\n",
    "import calendar\n",
    "import datetime\n",
    "import math"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:20:42.035153Z",
     "start_time": "2020-06-17T15:20:42.032622Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def mean0std1normalizer(df, features,stockname):\n",
    "    df = df[features]\n",
    "    mean_std_dict = {}\n",
    "    mu = df.mean()\n",
    "    std = df.std()\n",
    "    df = (df-mu)/std\n",
    "    mean_std_dict[stockname] = {'mean':mu,'std':std }\n",
    "\n",
    "    return df, mean_std_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-13T14:56:45.073225Z",
     "start_time": "2020-06-13T14:56:45.066474Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:20:42.349247Z",
     "start_time": "2020-06-17T15:20:42.344700Z"
    }
   },
   "outputs": [],
   "source": [
    "def zero_2_one_normalizerV2(df,features,name):\n",
    "\n",
    "    mean_std_dict = {}\n",
    "\n",
    "    ultimate = pd.DataFrame(columns=features)\n",
    "    \n",
    "    subdf = df[features]\n",
    "    max_ = subdf.max()\n",
    "    min_ = subdf.min()\n",
    "    subdf = ((subdf-min_) /(max_-min_))\n",
    "    ultimate = pd.concat([ultimate, subdf])\n",
    "    mean_std_dict[name] = {'min': min_, 'max': max_}\n",
    "\n",
    "    return ultimate, mean_std_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:20:42.491456Z",
     "start_time": "2020-06-17T15:20:42.488290Z"
    },
    "code_folding": [
     0
    ]
   },
   "outputs": [],
   "source": [
    "def volume_normalizer(df):\n",
    "    \n",
    "    stocklist = df['Name'].unique()\n",
    "    for name in stocklist:\n",
    "        subdf = df[df['Name'] == name]\n",
    "        sub = ((subdf['volume']-subdf['volume'].min()) /\n",
    "               (subdf['volume'].max()-subdf['volume'].min())).values\n",
    "        df.loc[(df['Name'] == name), 'volume'] = sub\n",
    "\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:20:42.649550Z",
     "start_time": "2020-06-17T15:20:42.641001Z"
    },
    "code_folding": [
     0
    ]
   },
   "outputs": [],
   "source": [
    "def gap_transform(df, features):\n",
    "\n",
    "    ultimate = pd.DataFrame(\n",
    "        columns='open high low close volume initopen name'.split())\n",
    "\n",
    "    stocklist = df['Name'].unique()\n",
    "\n",
    "    for name in stocklist:\n",
    "        subdf = df[df['Name'] == name]\n",
    "        initopen = list(subdf['initopen'].values).copy()\n",
    "        subdf = subdf[['open', 'high', 'low', 'close', 'volume']]\n",
    "        test = (subdf.iloc[1:].values-subdf.iloc[:-1].values)\n",
    "        test = pd.DataFrame(\n",
    "            test, columns=['open', 'high', 'low', 'close', 'volume'])\n",
    "        test['initopen'] = initopen[:-1]\n",
    "        test['name'] = name\n",
    "        ultimate = pd.concat([ultimate, test])\n",
    "\n",
    "    return ultimate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:20:42.901033Z",
     "start_time": "2020-06-17T15:20:42.897028Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def gap_transformV2(df, features):\n",
    "\n",
    "    subdf = df[features]\n",
    "    test = (subdf.iloc[1:].values-subdf.iloc[:-1].values)\n",
    "    test = pd.DataFrame(test, columns=features)\n",
    "\n",
    "    return test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:20:43.165486Z",
     "start_time": "2020-06-17T15:20:43.159571Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def column_extracter(df, src_feat, trg_feat):\n",
    "    '''\n",
    "    src_feat : a list containing feature name of the src\n",
    "    trg_feat : a list containing feature name of the trg\n",
    "    output : a dictionary with column name as key and np array as value\n",
    "    '''\n",
    "\n",
    "    src_buffer_dict, trg_buffer_dict, name_buffer_dict = {}, {} ,{}\n",
    "\n",
    "    for _ in src_feat:\n",
    "        \n",
    "\n",
    "        if df[_].dtype == object:\n",
    "\n",
    "            name_buffer_dict[_] = df[_].to_list()\n",
    "\n",
    "        else:\n",
    "\n",
    "            src_buffer_dict[_] = np.array(df[_].to_list(), dtype=np.float32)\n",
    "\n",
    "    for _ in trg_feat:\n",
    "\n",
    "        trg_buffer_dict[_] = np.array(df[_].to_list(), dtype=np.float32)\n",
    "\n",
    "    return src_buffer_dict, trg_buffer_dict, name_buffer_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:20:43.613946Z",
     "start_time": "2020-06-17T15:20:43.600784Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def srctrg_splitter(src_len, trg_len, srcdict, trgdict, skipsize, name):\n",
    "    \n",
    "    ### temporally get rid of open price with list index \n",
    "    \n",
    "    src_meta = np.vstack(\n",
    "        tuple(v for k, v in srcdict.items())).astype('float32')\n",
    "    trg_meta = np.vstack(\n",
    "        tuple(v for k, v in trgdict.items())).astype('float32')\n",
    "    \n",
    "    bufferdict = {'src': [], 'trg': [], 'initopen':[], 'name':[]}\n",
    "\n",
    "    i = 0\n",
    "    max_len = src_meta.shape[1]\n",
    "\n",
    "    while i+src_len+trg_len < max_len:\n",
    "#         src_ = src_meta[:-1, i:i+src_len]\n",
    "        realval = np.vstack((src_meta[:-1, i:i+src_len],np.zeros((1,src_len))))\n",
    "        print(realval)\n",
    "        src_ = np.hstack((np.full(\n",
    "            (src_meta.shape[0]-1, 1),-.1, dtype=np.float32),src_meta[:-1, i:i+src_len]))\n",
    "        trg_ = np.hstack((np.full(\n",
    "            (trg_meta.shape[0], 1),-.1, dtype=np.float32), trg_meta[:, i+src_len:i+src_len+trg_len]))\n",
    "#         trg_ =trg_meta[:, i+src_len-1:i+src_len+trg_len]\n",
    "        price_ = src_meta[-1, i:i+1]\n",
    "        bufferdict['src'].append(src_)\n",
    "        bufferdict['trg'].append(trg_)\n",
    "        bufferdict['initopen'].append(price_)\n",
    "        bufferdict['name'].append(str(name))\n",
    "        i += skipsize\n",
    "\n",
    "    output = pd.DataFrame(bufferdict)\n",
    "\n",
    "    return output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:20:44.116110Z",
     "start_time": "2020-06-17T15:20:44.111335Z"
    }
   },
   "outputs": [],
   "source": [
    "def srctrg_splitter_padded_trg(src_len, trg_len, srcdict, trgdict, skipsize, name):\n",
    "    \n",
    "    ### temporally get rid of open price with list index \n",
    "    \n",
    "    src_meta = np.vstack(\n",
    "        tuple(v for k, v in srcdict.items())).astype('float32')\n",
    "    trg_meta = np.vstack(\n",
    "        tuple(v for k, v in trgdict.items())).astype('float32')\n",
    "    \n",
    "    bufferdict = {'src': [], 'trg': [], 'initopen':[], 'name':[]}\n",
    "\n",
    "    i = 0\n",
    "    max_len = src_meta.shape[1]\n",
    "\n",
    "    while i+src_len+trg_len < max_len:\n",
    "\n",
    "\n",
    "        src_ = src_meta[:-1, i:i+src_len]\n",
    "        trg_ = trg_meta[:, i+src_len:i+src_len+trg_len]\n",
    "        price_ = src_meta[-1, i:i+1]\n",
    "        bufferdict['src'].append(src_)\n",
    "        bufferdict['trg'].append(trg_)\n",
    "        bufferdict['initopen'].append(price_)\n",
    "        bufferdict['name'].append(str(name))\n",
    "        i += skipsize\n",
    "\n",
    "    output = pd.DataFrame(bufferdict)\n",
    "\n",
    "    return output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:20:45.187137Z",
     "start_time": "2020-06-17T15:20:45.175972Z"
    }
   },
   "outputs": [],
   "source": [
    "def srctrg_splitter_withdummy(src_len, trg_len, srcdict, trgdict, skipsize, name):\n",
    "    \n",
    "    ### temporally get rid of open price with list index \n",
    "    \n",
    "    src_meta = np.vstack(\n",
    "        tuple(v for k, v in srcdict.items())).astype('float32')\n",
    "    trg_meta = np.vstack(\n",
    "        tuple(v for k, v in trgdict.items())).astype('float32')\n",
    "    \n",
    "    bufferdict = {'src': [], 'trg': [], 'initopen':[], 'name':[]}\n",
    "\n",
    "    i = 0\n",
    "    max_len = src_meta.shape[1]\n",
    "\n",
    "    while i+src_len+trg_len < max_len:\n",
    "\n",
    "        realval1 = np.vstack((src_meta[:-1, i:i+src_len],np.zeros((1,src_len), dtype=np.float32)))\n",
    "        realval1 = np.hstack(((np.ones((realval1.shape[0],1), dtype=np.float32)),realval1))\n",
    "        realval1[:-1,0]=0\n",
    "        \n",
    "        realval2 = np.vstack((trg_meta[:, i+src_len:i+src_len+trg_len],np.zeros((1,trg_len), dtype=np.float32)))\n",
    "        realval2 = np.hstack((np.ones((realval2.shape[0],1), dtype=np.float32),realval2))\n",
    "        realval2[:-1,0]=0\n",
    "\n",
    "\n",
    "#         src_ = src_meta[:-1, i:i+src_len]\n",
    "#         src_ = np.hstack((np.full(\n",
    "#             (src_meta.shape[0]-1, 1),-.1, dtype=np.float32),src_meta[:-1, i:i+src_len]))\n",
    "#         trg_ = np.hstack((np.full(\n",
    "#             (trg_meta.shape[0], 1),-.1, dtype=np.float32), trg_meta[:, i+src_len:i+src_len+trg_len]))\n",
    "#         trg_ =trg_meta[:, i+src_len-1:i+src_len+trg_len]\n",
    "        price_ = src_meta[-1, i:i+1]\n",
    "        bufferdict['src'].append(realval1)\n",
    "        bufferdict['trg'].append(realval2)\n",
    "        bufferdict['initopen'].append(price_)\n",
    "        bufferdict['name'].append(str(name))\n",
    "        i += skipsize\n",
    "\n",
    "    output = pd.DataFrame(bufferdict)\n",
    "\n",
    "    return output"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Process Blocks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-15T09:06:33.033892Z",
     "start_time": "2020-06-15T09:06:33.007348Z"
    }
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('techcompdata/5yrs/AAPL.csv')\n",
    "# df = df.round(2)\n",
    "# df = gap_transformV2(df,['Open','High','Low','Close','Volume'])\n",
    "# df=(df/df.std())\n",
    "df,dict_ = zero_2_one_normalizerV2(df,['Open','High','Low','Close','Volume'],'INTC')\n",
    "# df=(df/df.std())\n",
    "# df,dict_ = mean0std1normalizer(df,['Open','High','Low','Close','Volume'],'INTC')\n",
    "print(len(df))\n",
    "df.describe()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-15T09:06:33.885849Z",
     "start_time": "2020-06-15T09:06:33.405076Z"
    },
    "code_folding": [
     0
    ]
   },
   "outputs": [],
   "source": [
    "for e in ['Open','High','Low','Close','Volume']:\n",
    "    sns.distplot(df[e].values, kde=True, rug=False)\n",
    "    plt.title(e)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:27:45.795568Z",
     "start_time": "2020-06-17T15:27:45.568076Z"
    },
    "code_folding": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Open</th>\n",
       "      <th>Close</th>\n",
       "      <th>Volume</th>\n",
       "      <th>initopen</th>\n",
       "      <th>name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.139524</td>\n",
       "      <td>0.129577</td>\n",
       "      <td>0.156270</td>\n",
       "      <td>31.120001</td>\n",
       "      <td>INTC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.142823</td>\n",
       "      <td>0.135446</td>\n",
       "      <td>0.129861</td>\n",
       "      <td>31.260000</td>\n",
       "      <td>INTC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.154372</td>\n",
       "      <td>0.142723</td>\n",
       "      <td>0.201658</td>\n",
       "      <td>31.750000</td>\n",
       "      <td>INTC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.162149</td>\n",
       "      <td>0.152817</td>\n",
       "      <td>0.197392</td>\n",
       "      <td>32.080002</td>\n",
       "      <td>INTC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.169691</td>\n",
       "      <td>0.145540</td>\n",
       "      <td>0.299449</td>\n",
       "      <td>32.400002</td>\n",
       "      <td>INTC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>935</th>\n",
       "      <td>0.965231</td>\n",
       "      <td>0.989949</td>\n",
       "      <td>0.100628</td>\n",
       "      <td>365.000000</td>\n",
       "      <td>TTD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>936</th>\n",
       "      <td>0.929476</td>\n",
       "      <td>0.929874</td>\n",
       "      <td>0.142118</td>\n",
       "      <td>352.299988</td>\n",
       "      <td>TTD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>937</th>\n",
       "      <td>0.953970</td>\n",
       "      <td>0.949833</td>\n",
       "      <td>0.097391</td>\n",
       "      <td>361.000000</td>\n",
       "      <td>TTD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>938</th>\n",
       "      <td>0.910473</td>\n",
       "      <td>0.995905</td>\n",
       "      <td>0.165513</td>\n",
       "      <td>345.549988</td>\n",
       "      <td>TTD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>939</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.133779</td>\n",
       "      <td>377.350006</td>\n",
       "      <td>TTD</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>21070 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Open     Close    Volume    initopen  name\n",
       "0    0.139524  0.129577  0.156270   31.120001  INTC\n",
       "1    0.142823  0.135446  0.129861   31.260000  INTC\n",
       "2    0.154372  0.142723  0.201658   31.750000  INTC\n",
       "3    0.162149  0.152817  0.197392   32.080002  INTC\n",
       "4    0.169691  0.145540  0.299449   32.400002  INTC\n",
       "..        ...       ...       ...         ...   ...\n",
       "935  0.965231  0.989949  0.100628  365.000000   TTD\n",
       "936  0.929476  0.929874  0.142118  352.299988   TTD\n",
       "937  0.953970  0.949833  0.097391  361.000000   TTD\n",
       "938  0.910473  0.995905  0.165513  345.549988   TTD\n",
       "939  1.000000  1.000000  0.133779  377.350006   TTD\n",
       "\n",
       "[21070 rows x 5 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import glob\n",
    "l = glob.glob('./techcompdata/5yrs/raw_data/*')\n",
    "output=pd.DataFrame(columns=['Open','Close','Volume'])\n",
    "\n",
    "for e in l:\n",
    "    df = pd.read_csv(e)\n",
    "    newname = e.lstrip(r'./techcompdata/5yrs/raw_data/')\n",
    "    df2, _ = zero_2_one_normalizerV2(\n",
    "        df, ['Open','Close','Volume'],newname)\n",
    "#     df2 = gap_transformV2(df2,['Open','Close','Volume'])\n",
    "    # df2=(df2-df2.mean())/df2.std()\n",
    "    \n",
    "#     df2, _ = zero_2_one_normalizerV2(\n",
    "#         df, ['Open','Close','Volume'],newname)\n",
    "    # df2=df2*2-1\n",
    "    initopen = df['Open'].values\n",
    "    df2['initopen'] = initopen\n",
    "    df2['name'] = newname.rstrip(r'.csv')\n",
    "    output=pd.concat([df2,output])\n",
    "    # df2[(df2['Open']==0.0) | (df2['Volume']==0.0) | (df2['Close']==0.0)]\n",
    "    df2.to_csv('./techcompdata/5yrs/0to1standardized/'+newname,index=False)\n",
    "output\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:27:54.695244Z",
     "start_time": "2020-06-17T15:27:54.641161Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Open</th>\n",
       "      <th>Close</th>\n",
       "      <th>Volume</th>\n",
       "      <th>initopen</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>21070.000000</td>\n",
       "      <td>21070.000000</td>\n",
       "      <td>21070.000000</td>\n",
       "      <td>21070.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.391862</td>\n",
       "      <td>0.388478</td>\n",
       "      <td>0.134024</td>\n",
       "      <td>246.614082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.257564</td>\n",
       "      <td>0.257579</td>\n",
       "      <td>0.101814</td>\n",
       "      <td>371.072975</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.620000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.160683</td>\n",
       "      <td>0.156844</td>\n",
       "      <td>0.069374</td>\n",
       "      <td>59.345000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.363898</td>\n",
       "      <td>0.358703</td>\n",
       "      <td>0.107829</td>\n",
       "      <td>124.260002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.613961</td>\n",
       "      <td>0.610298</td>\n",
       "      <td>0.166087</td>\n",
       "      <td>207.550003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2645.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Open         Close        Volume      initopen\n",
       "count  21070.000000  21070.000000  21070.000000  21070.000000\n",
       "mean       0.391862      0.388478      0.134024    246.614082\n",
       "std        0.257564      0.257579      0.101814    371.072975\n",
       "min        0.000000      0.000000      0.000000      1.620000\n",
       "25%        0.160683      0.156844      0.069374     59.345000\n",
       "50%        0.363898      0.358703      0.107829    124.260002\n",
       "75%        0.613961      0.610298      0.166087    207.550003\n",
       "max        1.000000      1.000000      1.000000   2645.000000"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import glob\n",
    "l = glob.glob('./techcompdata/5yrs/0to1standardized/*')\n",
    "output=pd.DataFrame(columns='Open Close Volume\tinitopen'.split())\n",
    "for e in l:\n",
    "    df1=pd.read_csv(e)\n",
    "    output=pd.concat([df1,output])\n",
    "# output = (output-output.mean())/output.std()\n",
    "\n",
    "# output = output[['Open','Close', 'Volume']]\n",
    "# output = output/output.std()\n",
    "# output=output.round(3)\n",
    "output.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:27:58.593329Z",
     "start_time": "2020-06-17T15:27:58.581908Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Open</th>\n",
       "      <th>Close</th>\n",
       "      <th>Volume</th>\n",
       "      <th>initopen</th>\n",
       "      <th>name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.139524</td>\n",
       "      <td>0.129577</td>\n",
       "      <td>0.156270</td>\n",
       "      <td>31.120001</td>\n",
       "      <td>INTC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.142823</td>\n",
       "      <td>0.135446</td>\n",
       "      <td>0.129861</td>\n",
       "      <td>31.260000</td>\n",
       "      <td>INTC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.154372</td>\n",
       "      <td>0.142723</td>\n",
       "      <td>0.201658</td>\n",
       "      <td>31.750000</td>\n",
       "      <td>INTC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.162149</td>\n",
       "      <td>0.152817</td>\n",
       "      <td>0.197392</td>\n",
       "      <td>32.080002</td>\n",
       "      <td>INTC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.169691</td>\n",
       "      <td>0.145540</td>\n",
       "      <td>0.299449</td>\n",
       "      <td>32.400002</td>\n",
       "      <td>INTC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>935</th>\n",
       "      <td>0.965231</td>\n",
       "      <td>0.989949</td>\n",
       "      <td>0.100628</td>\n",
       "      <td>365.000000</td>\n",
       "      <td>TTD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>936</th>\n",
       "      <td>0.929476</td>\n",
       "      <td>0.929874</td>\n",
       "      <td>0.142118</td>\n",
       "      <td>352.299988</td>\n",
       "      <td>TTD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>937</th>\n",
       "      <td>0.953970</td>\n",
       "      <td>0.949833</td>\n",
       "      <td>0.097391</td>\n",
       "      <td>361.000000</td>\n",
       "      <td>TTD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>938</th>\n",
       "      <td>0.910473</td>\n",
       "      <td>0.995905</td>\n",
       "      <td>0.165513</td>\n",
       "      <td>345.549988</td>\n",
       "      <td>TTD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>939</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.133779</td>\n",
       "      <td>377.350006</td>\n",
       "      <td>TTD</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>21070 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Open     Close    Volume    initopen  name\n",
       "0    0.139524  0.129577  0.156270   31.120001  INTC\n",
       "1    0.142823  0.135446  0.129861   31.260000  INTC\n",
       "2    0.154372  0.142723  0.201658   31.750000  INTC\n",
       "3    0.162149  0.152817  0.197392   32.080002  INTC\n",
       "4    0.169691  0.145540  0.299449   32.400002  INTC\n",
       "..        ...       ...       ...         ...   ...\n",
       "935  0.965231  0.989949  0.100628  365.000000   TTD\n",
       "936  0.929476  0.929874  0.142118  352.299988   TTD\n",
       "937  0.953970  0.949833  0.097391  361.000000   TTD\n",
       "938  0.910473  0.995905  0.165513  345.549988   TTD\n",
       "939  1.000000  1.000000  0.133779  377.350006   TTD\n",
       "\n",
       "[21070 rows x 5 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:28:01.527601Z",
     "start_time": "2020-06-17T15:28:00.717469Z"
    },
    "code_folding": [
     0
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for e in ['Open','Close','Volume']:\n",
    "    sns.distplot(output[e].values, kde=True, rug=False,bins=200)\n",
    "    plt.title(e)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:39:39.634274Z",
     "start_time": "2020-06-17T15:39:39.523379Z"
    }
   },
   "outputs": [],
   "source": [
    "final = pd.DataFrame(columns='src trg initopen name'.split())\n",
    "for name in output['name'].unique():   \n",
    "    srcdict, trgdict, namedict = column_extracter(output[output['name']==name],\n",
    "                                                  ['Open','Close', 'Volume', 'initopen','name'],\n",
    "                                                  ['Open','Close', 'Volume'])\n",
    "    test = srctrg_splitter_padded_trg(90, 30, srcdict, trgdict, 1, name)\n",
    "    final = pd.concat([final,test])\n",
    "    final = shuffle(final)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:39:40.015260Z",
     "start_time": "2020-06-17T15:39:39.953592Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>src</th>\n",
       "      <th>trg</th>\n",
       "      <th>initopen</th>\n",
       "      <th>name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>273</th>\n",
       "      <td>[[0.028498704, 0.034398973, 0.033550564, 0.036...</td>\n",
       "      <td>[[0.07604798, 0.07759054, 0.08464771, 0.082372...</td>\n",
       "      <td>[97.39]</td>\n",
       "      <td>AAPL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72</th>\n",
       "      <td>[[0.11026423, 0.110193044, 0.110762894, 0.1287...</td>\n",
       "      <td>[[0.023221012, 0.0, 0.034831535, 0.025500378, ...</td>\n",
       "      <td>[69.09]</td>\n",
       "      <td>CRM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>426</th>\n",
       "      <td>[[0.34508634, 0.35905567, 0.34850243, 0.355212...</td>\n",
       "      <td>[[0.45671928, 0.4556213, 0.43902883, 0.4392118...</td>\n",
       "      <td>[133.6]</td>\n",
       "      <td>FB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>944</th>\n",
       "      <td>[[0.3780359, 0.38155577, 0.38683563, 0.437698,...</td>\n",
       "      <td>[[0.56353396, 0.5707497, 0.5739176, 0.5723337,...</td>\n",
       "      <td>[23.1]</td>\n",
       "      <td>AMD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>780</th>\n",
       "      <td>[[0.18345542, 0.18642256, 0.17759018, 0.186917...</td>\n",
       "      <td>[[0.23422998, 0.23539153, 0.22944577, 0.250342...</td>\n",
       "      <td>[301.84]</td>\n",
       "      <td>TSLA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>231</th>\n",
       "      <td>[[0.1304197, 0.12393632, 0.12358902, 0.1216787...</td>\n",
       "      <td>[[0.29493487, 0.2849204, 0.2849204, 0.29603475...</td>\n",
       "      <td>[79.83]</td>\n",
       "      <td>BABA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268</th>\n",
       "      <td>[[0.037813094, 0.041292015, 0.041800477, 0.038...</td>\n",
       "      <td>[[0.091870055, 0.09146864, 0.09093343, 0.08667...</td>\n",
       "      <td>[94.7]</td>\n",
       "      <td>NFLX</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>[[0.00791458, 0.0029355898, 0.006820932, 0.004...</td>\n",
       "      <td>[[0.041443627, 0.04055143, 0.037529502, 0.0395...</td>\n",
       "      <td>[22.0]</td>\n",
       "      <td>NVDA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1085</th>\n",
       "      <td>[[0.47888067, 0.48011264, 0.47395283, 0.478352...</td>\n",
       "      <td>[[0.91182685, 0.94280183, 0.938402, 0.99208033...</td>\n",
       "      <td>[28.83]</td>\n",
       "      <td>AMD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265</th>\n",
       "      <td>[[0.6542334, 0.6474829, 0.6550344, 0.66533184,...</td>\n",
       "      <td>[[0.6860412, 0.6677346, 0.72151035, 0.74794054...</td>\n",
       "      <td>[151.78]</td>\n",
       "      <td>IBM</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>19030 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    src  \\\n",
       "273   [[0.028498704, 0.034398973, 0.033550564, 0.036...   \n",
       "72    [[0.11026423, 0.110193044, 0.110762894, 0.1287...   \n",
       "426   [[0.34508634, 0.35905567, 0.34850243, 0.355212...   \n",
       "944   [[0.3780359, 0.38155577, 0.38683563, 0.437698,...   \n",
       "780   [[0.18345542, 0.18642256, 0.17759018, 0.186917...   \n",
       "...                                                 ...   \n",
       "231   [[0.1304197, 0.12393632, 0.12358902, 0.1216787...   \n",
       "268   [[0.037813094, 0.041292015, 0.041800477, 0.038...   \n",
       "48    [[0.00791458, 0.0029355898, 0.006820932, 0.004...   \n",
       "1085  [[0.47888067, 0.48011264, 0.47395283, 0.478352...   \n",
       "265   [[0.6542334, 0.6474829, 0.6550344, 0.66533184,...   \n",
       "\n",
       "                                                    trg  initopen  name  \n",
       "273   [[0.07604798, 0.07759054, 0.08464771, 0.082372...   [97.39]  AAPL  \n",
       "72    [[0.023221012, 0.0, 0.034831535, 0.025500378, ...   [69.09]   CRM  \n",
       "426   [[0.45671928, 0.4556213, 0.43902883, 0.4392118...   [133.6]    FB  \n",
       "944   [[0.56353396, 0.5707497, 0.5739176, 0.5723337,...    [23.1]   AMD  \n",
       "780   [[0.23422998, 0.23539153, 0.22944577, 0.250342...  [301.84]  TSLA  \n",
       "...                                                 ...       ...   ...  \n",
       "231   [[0.29493487, 0.2849204, 0.2849204, 0.29603475...   [79.83]  BABA  \n",
       "268   [[0.091870055, 0.09146864, 0.09093343, 0.08667...    [94.7]  NFLX  \n",
       "48    [[0.041443627, 0.04055143, 0.037529502, 0.0395...    [22.0]  NVDA  \n",
       "1085  [[0.91182685, 0.94280183, 0.938402, 0.99208033...   [28.83]   AMD  \n",
       "265   [[0.6860412, 0.6677346, 0.72151035, 0.74794054...  [151.78]   IBM  \n",
       "\n",
       "[19030 rows x 4 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:28:32.894880Z",
     "start_time": "2020-06-17T15:28:32.584591Z"
    }
   },
   "outputs": [],
   "source": [
    "final.to_pickle('techcompdata/5yrs/srctrg_set90_30VOC_trgpad2.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-17T15:39:46.189706Z",
     "start_time": "2020-06-17T15:39:46.043876Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>src</th>\n",
       "      <th>trg</th>\n",
       "      <th>initopen</th>\n",
       "      <th>name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1026</th>\n",
       "      <td>[[0.5609239, 0.58449215, 0.582371, 0.572708, 0...</td>\n",
       "      <td>[[0.76596755, 0.78435075, 0.7791657, 0.7610182...</td>\n",
       "      <td>[49.0]</td>\n",
       "      <td>INTC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>643</th>\n",
       "      <td>[[0.6139815, 0.6396023, 0.6580247, 0.6622338, ...</td>\n",
       "      <td>[[0.6577198, 0.67516625, 0.6579028, 0.6507045,...</td>\n",
       "      <td>[177.68]</td>\n",
       "      <td>FB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>[[0.5802059, 0.5314646, 0.5860413, 0.62311214,...</td>\n",
       "      <td>[[0.3478261, 0.27665904, 0.28215107, 0.3115561...</td>\n",
       "      <td>[145.31]</td>\n",
       "      <td>IBM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>396</th>\n",
       "      <td>[[0.16833645, 0.16770604, 0.16638224, 0.169376...</td>\n",
       "      <td>[[0.23443726, 0.2425603, 0.24308261, 0.2480446...</td>\n",
       "      <td>[798.0]</td>\n",
       "      <td>AMZN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1063</th>\n",
       "      <td>[[0.4609697, 0.47132146, 0.46540612, 0.4820957...</td>\n",
       "      <td>[[0.5851907, 0.5874089, 0.61244327, 0.6143446,...</td>\n",
       "      <td>[98.03]</td>\n",
       "      <td>EA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>745</th>\n",
       "      <td>[[0.16507767, 0.17483008, 0.17869425, 0.181914...</td>\n",
       "      <td>[[0.12987475, 0.14063917, 0.13249686, 0.136487...</td>\n",
       "      <td>[285.86]</td>\n",
       "      <td>TSLA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>426</th>\n",
       "      <td>[[0.17905207, 0.18271568, 0.17704679, 0.181790...</td>\n",
       "      <td>[[0.20998032, 0.2116386, 0.20704949, 0.2044657...</td>\n",
       "      <td>[136.43]</td>\n",
       "      <td>AAPL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>[[0.0054558255, 0.009855685, 0.008623724, 0.00...</td>\n",
       "      <td>[[0.013199578, 0.013023583, 0.013903555, 0.013...</td>\n",
       "      <td>[1.93]</td>\n",
       "      <td>AMD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>265</th>\n",
       "      <td>[[0.07783515, 0.0814003, 0.086805515, 0.089289...</td>\n",
       "      <td>[[0.056041766, 0.04820995, 0.052534122, 0.0465...</td>\n",
       "      <td>[210.0]</td>\n",
       "      <td>TSLA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338</th>\n",
       "      <td>[[0.13497958, 0.13449031, 0.13604443, 0.135008...</td>\n",
       "      <td>[[0.22396824, 0.23873252, 0.2462154, 0.2433085...</td>\n",
       "      <td>[66.15]</td>\n",
       "      <td>NVDA</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>19030 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    src  \\\n",
       "1026  [[0.5609239, 0.58449215, 0.582371, 0.572708, 0...   \n",
       "643   [[0.6139815, 0.6396023, 0.6580247, 0.6622338, ...   \n",
       "76    [[0.5802059, 0.5314646, 0.5860413, 0.62311214,...   \n",
       "396   [[0.16833645, 0.16770604, 0.16638224, 0.169376...   \n",
       "1063  [[0.4609697, 0.47132146, 0.46540612, 0.4820957...   \n",
       "...                                                 ...   \n",
       "745   [[0.16507767, 0.17483008, 0.17869425, 0.181914...   \n",
       "426   [[0.17905207, 0.18271568, 0.17704679, 0.181790...   \n",
       "34    [[0.0054558255, 0.009855685, 0.008623724, 0.00...   \n",
       "265   [[0.07783515, 0.0814003, 0.086805515, 0.089289...   \n",
       "338   [[0.13497958, 0.13449031, 0.13604443, 0.135008...   \n",
       "\n",
       "                                                    trg  initopen  name  \n",
       "1026  [[0.76596755, 0.78435075, 0.7791657, 0.7610182...    [49.0]  INTC  \n",
       "643   [[0.6577198, 0.67516625, 0.6579028, 0.6507045,...  [177.68]    FB  \n",
       "76    [[0.3478261, 0.27665904, 0.28215107, 0.3115561...  [145.31]   IBM  \n",
       "396   [[0.23443726, 0.2425603, 0.24308261, 0.2480446...   [798.0]  AMZN  \n",
       "1063  [[0.5851907, 0.5874089, 0.61244327, 0.6143446,...   [98.03]    EA  \n",
       "...                                                 ...       ...   ...  \n",
       "745   [[0.12987475, 0.14063917, 0.13249686, 0.136487...  [285.86]  TSLA  \n",
       "426   [[0.20998032, 0.2116386, 0.20704949, 0.2044657...  [136.43]  AAPL  \n",
       "34    [[0.013199578, 0.013023583, 0.013903555, 0.013...    [1.93]   AMD  \n",
       "265   [[0.056041766, 0.04820995, 0.052534122, 0.0465...   [210.0]  TSLA  \n",
       "338   [[0.22396824, 0.23873252, 0.2462154, 0.2433085...   [66.15]  NVDA  \n",
       "\n",
       "[19030 rows x 4 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.read_pickle('techcompdata/5yrs/srctrg_set90_30VOC_trgpad2.pkl')\n",
    "# df=shuffle(df)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-15T11:38:07.460487Z",
     "start_time": "2020-06-15T11:38:07.443636Z"
    }
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('techcompdata/IBM.csv')\n",
    "df = gap_transformV2(df,['Open','High','Low','Close','Volume'])\n",
    "stdrecord.loc['IBM']=df.std().values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-13T16:29:56.905341Z",
     "start_time": "2020-06-13T16:29:56.896499Z"
    }
   },
   "outputs": [],
   "source": [
    "stdrecord=pd.DataFrame(columns=['Open','High','Low','Close','Volume'],\n",
    "                      index='IBM AMD INTC'.split())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-13T16:49:53.285124Z",
     "start_time": "2020-06-13T16:49:53.278644Z"
    }
   },
   "outputs": [],
   "source": [
    "stdrecord.to_csv('techcompdata/stdrecord.csv',)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-13T16:49:54.152555Z",
     "start_time": "2020-06-13T16:49:54.140971Z"
    }
   },
   "outputs": [],
   "source": [
    "pd.read_csv('techcompdata/stdrecord.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-13T16:50:01.845798Z",
     "start_time": "2020-06-13T16:50:01.837488Z"
    }
   },
   "outputs": [],
   "source": [
    "stdrecord"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}