{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "def initialize_lists(size=30):\n",
    "    initial_values = [0] * size\n",
    "    return initial_values.copy(), initial_values.copy(), initial_values.copy()\n",
    "\n",
    "def get_window(df, len_df):\n",
    "    if len_df > 30:\n",
    "        return df[len_df-31:len_df].astype('float32')\n",
    "    else:\n",
    "        return None\n",
    "\n",
    "def transform_window(df_window, scaler):\n",
    "    return scaler.transform(df_window)\n",
    "\n",
    "def prepare_input(df_trans):\n",
    "    return df_trans[:30,:].reshape((1,30,30))\n",
    "\n",
    "def predict(model, df_new):\n",
    "    return model.predict(df_new)\n",
    "\n",
    "def calculate_residuals(df_trans, pred):\n",
    "    actual = df_trans[30,:25]\n",
    "    resid = actual - pred\n",
    "    return actual, resid\n",
    "\n",
    "def resize_prediction(pred, df_trans):\n",
    "    pred.resize((pred.shape[0], pred.shape[1] + len(df_trans[30,25:])))\n",
    "    pred[:, -len(df_trans[30,25:]):] = df_trans[30,25:]\n",
    "    return pred\n",
    "\n",
    "def inverse_transform(scaler, pred, df_trans):\n",
    "    pred = scaler.inverse_transform(np.array(pred))\n",
    "    actual = scaler.inverse_transform(np.array([df_trans[30,:]]))\n",
    "    return actual, pred\n",
    "\n",
    "def update_lists(actual_list, pred_list, resid_list, actual, pred, resid):\n",
    "    actual_list.pop(0)\n",
    "    pred_list.pop(0)\n",
    "    resid_list.pop(0)\n",
    "    actual_list.append(actual[0,1])\n",
    "    pred_list.append(pred[0,1])\n",
    "    resid_list.append(resid[0,1])\n",
    "    return actual_list, pred_list, resid_list\n",
    "\n",
    "def calculate_distances(resid, kmeans1, kmeans2, kmeans3, kmeans4):\n",
    "    dist = []\n",
    "    dist.append(np.linalg.norm(resid[:,1:8]-kmeans1.cluster_centers_[0], ord=2, axis=1))\n",
    "    dist.append(np.linalg.norm(resid[:,8:15]-kmeans2.cluster_centers_[0], ord=2, axis=1))\n",
    "    dist.append(np.linalg.norm(resid[:,15:22]-kmeans3.cluster_centers_[0], ord=2, axis=1))\n",
    "    dist.append(np.linalg.norm(resid[:,22:29]-kmeans4.cluster_centers_[0], ord=2, axis=1))\n",
    "    return np.array(dist)\n",
    "\n",
    "def pipeline(df, scaler, model, kmeans1, kmeans2, kmeans3, kmeans4):\n",
    "    actual_list, pred_list, resid_list = initialize_lists()\n",
    "    len_df = np.len(df)\n",
    "    df_window = get_window(df, len_df)\n",
    "    if df_window is not None:\n",
    "        df_trans = transform_window(df_window, scaler)\n",
    "        df_new = prepare_input(df_trans)\n",
    "        pred = predict(model, df_new)\n",
    "        actual, resid = calculate_residuals(df_trans, pred)\n",
    "        pred = resize_prediction(pred, df_trans)\n",
    "        actual, pred = inverse_transform(scaler, pred, df_trans)\n",
    "        actual_list, pred_list, resid_list = update_lists(actual_list, pred_list, resid_list, actual, pred, resid)\n",
    "        dist = calculate_distances(resid, kmeans1, kmeans2, kmeans3, kmeans4)\n",
    "        return actual_list, pred_list, resid_list, dist\n",
    "    else:\n",
    "        return actual_list, pred_list, resid_list, None\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['a', 'b', 'c', 'd']"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inp = ['a', 'b']\n",
    "out = ['c','d']\n",
    "\n",
    "inp+out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "smartbuilding",
   "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": "3.11.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}