{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2288179b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 768/768 [01:11<00:00, 10.72it/s]\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "from glob import glob\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.decomposition import PCA, KernelPCA\n",
    "from sklearn.preprocessing import (StandardScaler, \n",
    "                                   MinMaxScaler, \n",
    "                                   MaxAbsScaler,\n",
    "                                   PowerTransformer,\n",
    "                                   Binarizer)\n",
    "\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.model_selection import cross_validate\n",
    "from sklearn.metrics import classification_report, accuracy_score\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from math import isqrt\n",
    "import pickle\n",
    "from tqdm import tqdm\n",
    "import os\n",
    "\n",
    "os.environ['TF_FORCE_GPU_ALLOW_GROWTH'] = 'true'\n",
    "os.environ['CUDA_VISIBLE_DEVICES'] = '2'\n",
    "\n",
    "def load_data(user_filter=None):\n",
    "    dic_data = []\n",
    "    \n",
    "    for p in tqdm(glob('/opt/iui-datarelease3-sose2021/*.csv')):\n",
    "        path = p\n",
    "        filename = path.split('/')[-1]\n",
    "        user = int(filename.split('_')[0][1:])\n",
    "        if (user_filter):\n",
    "            if (user != user_filter):\n",
    "                continue\n",
    "        scenario = filename.split('_')[1][len('Scenario'):]\n",
    "        heightnorm = filename.split('_')[2][len('HeightNormalization'):] == 'True'\n",
    "        armnorm = filename.split('_')[3][len('ArmNormalization'):] == 'True'\n",
    "        rep =  int(filename.split('.')[0].split('_')[4][len('Repetition'):])\n",
    "        session = filename.split('_')[5][len('Session'):]\n",
    "        session = session.split('.')[0]\n",
    "      \n",
    "        data = pd.read_csv(path)\n",
    "        dic_data.append(\n",
    "            {\n",
    "                'filename': path,\n",
    "                'user': user,\n",
    "                'scenario': scenario,\n",
    "                'heightnorm': heightnorm,\n",
    "                'armnorm': armnorm,\n",
    "                'rep': rep,\n",
    "                'session': session,\n",
    "                'data': data \n",
    "                \n",
    "            }\n",
    "        )\n",
    "    return dic_data\n",
    "\n",
    "dic_data = load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "3df066af",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# dataP = pd.DataFrame.from_dict(fil_dic_data) #pandas Dataframe Form mit 'data0' nur die daten\n",
    "\n",
    "# tempP = dataP['data']\n",
    "\n",
    "# tempP = tempP[0].drop(columns=['Scenario','HeightNormalization','ArmNormalization','LeftHandTrackingAccuracy','RightHandTrackingAccuracy']) #P without String Data\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "30296cad",
   "metadata": {},
   "outputs": [],
   "source": [
    "fil_dic_data = []\n",
    "for d in dic_data:\n",
    "    if d['scenario'] == 'Sorting':\n",
    "        if d['heightnorm'] == d['armnorm']:\n",
    "            fil_dic_data.append(d)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a808f50",
   "metadata": {},
   "source": [
    "Test\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "dc206ded",
   "metadata": {},
   "outputs": [],
   "source": [
    "min_Max   = MinMaxScaler()\n",
    "standard  = StandardScaler()\n",
    "max_Abs   = MaxAbsScaler()\n",
    "binarizer = Binarizer()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "cf5b5695",
   "metadata": {},
   "outputs": [],
   "source": [
    "session_data_1 = []\n",
    "session_data_2 = []\n",
    "\n",
    "user_data_1    = []\n",
    "user_data_2    = []\n",
    "\n",
    "data_1         = []\n",
    "data_2         = []\n",
    "\n",
    "for a in fil_dic_data:\n",
    "    if(a['session'] == '1'): ## Daten aus session 1 für train\n",
    "        session_data_1.append(a)\n",
    "        \n",
    "    if(a['session'] == '2'): ## Daten aus Session 2 zum validaten\n",
    "        session_data_2.append(a)\n",
    "\n",
    "for b in session_data_1:\n",
    "    user_data_1.append(b['user']) ## Label zu 1 \n",
    "    data_1.append(a['data'])\n",
    "    \n",
    "for c in session_data_2:\n",
    "    user_data_2.append(b['user']) ## Label zu 2\n",
    "    data_2.append(a['data'])\n",
    "    \n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "82465bca",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 96/96 [00:00<00:00, 575.65it/s]\n",
      "100%|██████████| 96/96 [00:00<00:00, 646.85it/s]\n"
     ]
    }
   ],
   "source": [
    "dataF_1 = [] ## Filtered Data session 1\n",
    "dataF_2 = [] ## Filtered Data session 2\n",
    "\n",
    "temp_1  = [] ## Temp Holder für 1\n",
    "temp_2  = [] ## Temp Holder für 2\n",
    "\n",
    "counter = 0 ## Counter für Einspeisen der Daten\n",
    "\n",
    "\n",
    "for a in data_1:\n",
    "    temp_1.append(pd.DataFrame(a))\n",
    "\n",
    "for b in data_2:\n",
    "    temp_2.append(pd.DataFrame(b))\n",
    "\n",
    "\n",
    "for c in tqdm(temp_1):\n",
    "    dataF_1.append(c.drop(columns=['Scenario','HeightNormalization','ArmNormalization','LeftHandTrackingAccuracy','RightHandTrackingAccuracy','Unnamed: 0', 'FrameID','participantID','Repetition']))\n",
    "    counter +=1\n",
    "    \n",
    "counter = 0\n",
    "\n",
    "for d in tqdm(temp_2):\n",
    "    dataF_2.append(c.drop(columns=['Scenario','HeightNormalization','ArmNormalization','LeftHandTrackingAccuracy','RightHandTrackingAccuracy','Unnamed: 0', 'FrameID','participantID','Repetition']))\n",
    "    counter +=1\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "a6f7076e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 96/96 [00:03<00:00, 25.87it/s]\n"
     ]
    }
   ],
   "source": [
    "minD = [] ## normalisierte Daten durch Minmax\n",
    "staD = [] ## normalisierte Daten durch Standard\n",
    "maxD = [] ## normalisierte Daten durch MaxAbs\n",
    "binD = [] ## normalisierte Daten durch binarizer\n",
    "\n",
    "for i in tqdm(dataF_1):\n",
    "    minD.append( min_Max.fit_transform(i))\n",
    "    staD.append(standard.fit_transform(i))\n",
    "    maxD.append(max_Abs.fit_transform(i))\n",
    "    binD.append(binarizer.fit_transform(i))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "94676652",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "x_train,x_test,y_train,y_test = train_test_split(minD,user_data_1,random_state=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14a4abe1",
   "metadata": {},
   "source": [
    "Classi"
   ]
  }
 ],
 "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}