{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from skimage import img_as_ubyte\n",
    "from fileSetUtils import *\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "marker_dataset = decompress(\"data/set4.zip\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pre Processing Data\n",
    "\n",
    "- Convert to GrayScale Array\n",
    "- Add brightness Variants\n",
    "- Match Postiv and negativ sampeles\n",
    "- Add Image Noise\n",
    "- Convert to Vector"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                  images  label         name  \\\n",
      "768    [[0.11547650284867754, 0.10604260082346613, 0....      1  P_65079.png   \n",
      "38165  [[0.2675255918688092, 0.25940904533106907, 0.2...      1  P_65008.png   \n",
      "826    [[0.3573983704635319, 0.3524469445043463, 0.35...      1  P_65315.png   \n",
      "45882  [[0.3036945143677907, 0.33255084627388065, 0.3...      1  P_64893.png   \n",
      "5079   [[0.6330085475932614, 0.6413172860999162, 0.63...      1  P_64956.png   \n",
      "27769  [[0.3966944347100451, 0.3904759700894089, 0.40...      1  P_65307.png   \n",
      "42321  [[1.0, 0.7106636902804104, 0.4187904423537007,...      1  P_65090.png   \n",
      "16995  [[0.3847586898669263, 0.4049870069217676, 0.42...      1  P_65381.png   \n",
      "53259  [[0.2213170565795373, 0.22499321240258785, 0.2...     -1  N_43926.png   \n",
      "11090  [[0.29309898571769444, 0.2946742023583296, 0.2...     -1  N_19987.png   \n",
      "48972  [[0.5295921555981438, 0.5322508742486868, 0.56...     -1  N_59614.png   \n",
      "57798  [[0.3524123259285395, 0.3578618893750531, 0.36...     -1  N_45830.png   \n",
      "49024  [[0.5713109358296202, 0.5766120227369558, 0.58...     -1  N_12918.png   \n",
      "37182  [[0.5989643653604785, 0.5852596222729902, 0.59...     -1  N_42979.png   \n",
      "65168  [[1.0, 1.0, 1.0, 0.9103408074504593, 0.8179134...     -1  N_34581.png   \n",
      "10195  [[0.9891486359078651, 0.9973245534560057, 0.99...     -1  N_48983.png   \n",
      "\n",
      "       variant     id  \n",
      "768          1    768  \n",
      "38165        1  38165  \n",
      "826          2    826  \n",
      "45882        2  45882  \n",
      "5079         3   5079  \n",
      "27769        3  27769  \n",
      "42321        4  42321  \n",
      "16995        4  16995  \n",
      "53259        1  53259  \n",
      "11090        1  11090  \n",
      "48972        2  48972  \n",
      "57798        2  57798  \n",
      "49024        3  49024  \n",
      "37182        3  37182  \n",
      "65168        4  65168  \n",
      "10195        4  10195  \n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 16 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "New: 129756 (+64878 -64878, 0.50% 0.50%) | Old: 65451 (+573 -64878, 0.01% 0.99%)\n"
     ]
    },
    {
     "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>images</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>23290</th>\n",
       "      <td>[[0.40643760463074197, 0.3982520862496269, 0.4...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47604</th>\n",
       "      <td>[[0.12182608906416852, 0.19741380979252504, 0....</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54068</th>\n",
       "      <td>[[0.5783613612865524, 0.4507420836595933, 0.47...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43583</th>\n",
       "      <td>[[0.31308472117490954, 0.30893375779528137, 0....</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56001</th>\n",
       "      <td>[[0.2915222336036848, 0.27913560835588913, 0.2...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                  images  label\n",
       "23290  [[0.40643760463074197, 0.3982520862496269, 0.4...      1\n",
       "47604  [[0.12182608906416852, 0.19741380979252504, 0....      1\n",
       "54068  [[0.5783613612865524, 0.4507420836595933, 0.47...      1\n",
       "43583  [[0.31308472117490954, 0.30893375779528137, 0....      1\n",
       "56001  [[0.2915222336036848, 0.27913560835588913, 0.2...      1"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Pipeline Functions\n",
    "def add_additional_col(df: pd.DataFrame) -> pd.DataFrame:\n",
    "    df[\"variant\"] = 0\n",
    "    df[\"id\"] = df.index\n",
    "    return df\n",
    "\n",
    "def clean_up_image_array(df: pd.DataFrame) -> pd.DataFrame:\n",
    "    df[\"images\"] = df[\"images\"].apply(lambda x: x[:, :, 0].astype(np.float32) / 255)\n",
    "    return df\n",
    "\n",
    "def add_var_brightness(df: pd.DataFrame, levels: list[int]) -> pd.DataFrame:\n",
    "    new_dfs = [df]\n",
    "    for n,l in enumerate(levels):\n",
    "        new_df = df.query(\"variant == 0\").copy()\n",
    "        new_df[\"images\"] = new_df[\"images\"].apply(lambda x: np.clip(x*l, a_min=0.0, a_max=1.0))\n",
    "        new_df[\"variant\"] = n+1\n",
    "        new_dfs.append(new_df)\n",
    "\n",
    "    return pd.concat(new_dfs)\n",
    "\n",
    "def match_sample_size(df: pd.DataFrame, size: int) -> pd.DataFrame:\n",
    "    df_pos = df[df.label ==  1].sample(size//2, replace=True)\n",
    "    df_neg = df[df.label == -1].sample(size//2, replace=True)\n",
    "    return pd.concat([df_pos, df_neg])\n",
    "\n",
    "def add_noise(df: pd.DataFrame, level) -> pd.DataFrame:\n",
    "    df[\"images\"] = df[\"images\"].apply(lambda x: np.clip(x + np.random.normal(0, level, x.shape), a_min=0.0, a_max=1.0))\n",
    "    return df\n",
    "\n",
    "def remove_uncessary_cols(df: pd.DataFrame) -> pd.DataFrame:\n",
    "    return df.drop(columns=[\"id\", \"variant\", \"name\"])\n",
    "\n",
    "def change_label_values(df: pd.DataFrame) -> pd.DataFrame:\n",
    "    #df[\"label\"] = df[\"label\"].apply(lambda x: np.array([1, 0]) if x == 1 else np.array([0, 1]))\n",
    "    df[\"label\"] = df[\"label\"].apply(lambda x: 1 if x == 1 else 0)\n",
    "    return df\n",
    "\n",
    "def vectorize_images(df: pd.DataFrame) -> pd.DataFrame:\n",
    "    #df[\"images\"] = df[\"images\"].apply(lambda x: np.expand_dims(x, axis=2))\n",
    "    return df\n",
    "\n",
    "# Collecting before metrics\n",
    "size_old = marker_dataset.shape[0]\n",
    "size_old_pos = marker_dataset[marker_dataset.label == 1].shape[0]\n",
    "size_old_neg = marker_dataset[marker_dataset.label == -1].shape[0]\n",
    "\n",
    "# Running the main pipeline\n",
    "marker_dataset_mod = (marker_dataset.copy()\n",
    "    .pipe(add_additional_col)\n",
    "    .pipe(clean_up_image_array)\n",
    "    .pipe(add_var_brightness, levels=[0.5, 0.75, 1.25, 1.5])\n",
    "    .pipe(match_sample_size, size=size_old_neg*2)\n",
    "    .pipe(add_noise, level=0.01)\n",
    ")\n",
    "\n",
    "# Plotting Sample Data\n",
    "samples = pd.concat([\n",
    "    marker_dataset_mod[(marker_dataset_mod.label ==  1) & (marker_dataset_mod.variant == 1)].sample(2, replace=True, random_state=42),\n",
    "    marker_dataset_mod[(marker_dataset_mod.label ==  1) & (marker_dataset_mod.variant == 2)].sample(2, replace=True, random_state=42),\n",
    "    marker_dataset_mod[(marker_dataset_mod.label ==  1) & (marker_dataset_mod.variant == 3)].sample(2, replace=True, random_state=42),\n",
    "    marker_dataset_mod[(marker_dataset_mod.label ==  1) & (marker_dataset_mod.variant == 4)].sample(2, replace=True, random_state=42),\n",
    "    marker_dataset_mod[(marker_dataset_mod.label == -1) & (marker_dataset_mod.variant == 1)].sample(2, replace=True, random_state=42),\n",
    "    marker_dataset_mod[(marker_dataset_mod.label == -1) & (marker_dataset_mod.variant == 2)].sample(2, replace=True, random_state=42),\n",
    "    marker_dataset_mod[(marker_dataset_mod.label == -1) & (marker_dataset_mod.variant == 3)].sample(2, replace=True, random_state=42),\n",
    "    marker_dataset_mod[(marker_dataset_mod.label == -1) & (marker_dataset_mod.variant == 4)].sample(2, replace=True, random_state=42)\n",
    "])\n",
    "print(samples)\n",
    "fig = plt.figure(figsize=(8, 8))\n",
    "for i in range(16):\n",
    "    s = samples.iloc[i]\n",
    "    ax = fig.add_subplot(4, 4, i+1)\n",
    "    ax.set_title(\"{}-V{} ({})\".format(\n",
    "        s[\"id\"], \n",
    "        s[\"variant\"],\n",
    "        s[\"label\"]\n",
    "    ))\n",
    "    plt.imshow(s[\"images\"], cmap=\"gray\",  vmin=0.0, vmax=1.0)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "# Metrics\n",
    "size_new = marker_dataset_mod.shape[0]\n",
    "size_new_pos = marker_dataset_mod[marker_dataset_mod.label == 1].shape[0]\n",
    "size_new_neg = marker_dataset_mod[marker_dataset_mod.label == -1].shape[0]\n",
    "\n",
    "# Finalize Data Set\n",
    "marker_dataset_mod = (marker_dataset_mod\n",
    "    .pipe(remove_uncessary_cols)\n",
    "    .pipe(vectorize_images)\n",
    "    .pipe(change_label_values)\n",
    ")\n",
    "\n",
    "print(\"New: {} (+{} -{}, {:.02f}% {:.02f}%) | Old: {} (+{} -{}, {:.02f}% {:.02f}%)\".format(\n",
    "    size_new, size_new_pos, size_new_neg, size_new_pos/size_new, size_new_neg/size_new,\n",
    "    size_old, size_old_pos, size_old_neg, size_old_pos/size_old, size_old_neg/size_old,\n",
    "))\n",
    "marker_dataset_mod.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save Data Set\n",
    "marker_dataset_mod.to_pickle(\"data/set4.pkl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}