test copy.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "a\n"
     ]
    }
   ],
   "source": [
    "if not 0%2:\n",
    "    print('a')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "\n",
    "override_image_path = True\n",
    "override_image_path_string = '/data_store/faces'\n",
    "override_leave_parent_level = ''\n",
    "\n",
    "ss = '/media/gamal/Passport/Datasets/VoxCeleb2/Voxceleb2Faces/dev/mp4/id00012/2DLq_Kkc1r8/1.png'\n",
    "\n",
    "ssn = '/data_store/faces/dev/mp4/id00012/2DLq_Kkc1r8/1.png'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/media/gamal/Passport/Datasets/VoxCeleb2/Voxceleb2Faces'"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "\n",
    "def get_parent_by_level(file,parent_level):\n",
    "    path = Path(file)\n",
    "    file = path.parent.absolute()\n",
    "    if(parent_level == 0):\n",
    "        return str(file)\n",
    "    else:\n",
    "        return get_parent_by_level(file,parent_level - 1)\n",
    "\n",
    "xx = get_parent_by_level(ss,4)\n",
    "xx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/data_store/faces/dev/mp4/id00012/2DLq_Kkc1r8/1.png'"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ss.replace( xx, override_image_path_string)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "ename": "IndentationError",
     "evalue": "expected an indented block (470112864.py, line 4)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;36m  Input \u001b[0;32mIn [2]\u001b[0;36m\u001b[0m\n\u001b[0;31m    else:\u001b[0m\n\u001b[0m    ^\u001b[0m\n\u001b[0;31mIndentationError\u001b[0m\u001b[0;31m:\u001b[0m expected an indented block\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "def change_image_path(og_path):\n",
    "    if(override_image_path):\n",
    "        pass\n",
    "    else:\n",
    "        return og_path\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('/data_store/checkpoint/noablation/loss_total_2.picke', 'rb') as handle:\n",
    "    loss_list = pickle.load(handle)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "20"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(loss_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[3185, 0.8733890643343329],\n",
       " [3791, 0.8858071453869343],\n",
       " [4369, 0.8849454987794161],\n",
       " [4961, 0.8971156887710094],\n",
       " [5491, 0.8993238937109709],\n",
       " [5848, 0.8746841084212065],\n",
       " [6027, 0.8944266103208065],\n",
       " [6192, 0.8627886716276407],\n",
       " [6610, 0.8595869988203049],\n",
       " [6835, 0.8109119478613138],\n",
       " [6992, 0.7945402376353741],\n",
       " [7152, 0.7643793243914843],\n",
       " [7311, 0.7225744239985943],\n",
       " [7474, 0.6597365373745561],\n",
       " [7634, 0.6305282451212406],\n",
       " [7786, 0.5898538427427411],\n",
       " [7937, 0.5298748090863228],\n",
       " [8057, 0.4713828880339861],\n",
       " [8302, 0.4713828880339861]]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "loss_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2022027"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "last_sm = loss_list[-1:][0][0]\n",
    "last_sm\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "del loss_list[-60:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[2021853, 0.01830843836069107],\n",
       " [2021873, 0.017935167823452502],\n",
       " [2021893, 0.01774855743860826],\n",
       " [2021912, 0.01989271049387753],\n",
       " [2021931, 0.019403521728236228],\n",
       " [2021951, 0.01631369796814397],\n",
       " [2021970, 0.018611144507303834],\n",
       " [2021990, 0.018678128602914512],\n",
       " [2022008, 0.022129257617052644],\n",
       " [2022027, 0.018837745185010135]]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "loss_list[-10::]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('/data_store/checkpoint/noablation/correct.picke', 'wb') as handle:\n",
    "    pickle.dump(loss_list, handle)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('/data_store/checkpoint/noablation/last_sample_no1.picke', 'rb') as handle:\n",
    "    ls_sm_og = pickle.load(handle)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2022027"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ls_sm_og"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('/data_store/checkpoint/noablation/last_sample_no1.picke', 'wb') as handle:\n",
    "    pickle.dump(last_sm, handle)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "22"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "6689 - 6667"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[762817, 0.01910052530001849]]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "loss_list[6666::]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_val = [x[1] for x in loss_list]\n",
    "x_val = [x[0] for x in loss_list]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "loss_list = y_val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1000 1001 1002 ... 6649 6650 6651]\n"
     ]
    }
   ],
   "source": [
    "from scipy.signal import savgol_filter\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import math\n",
    "\n",
    "plt.plot(x_val[1000::],loss_list[1000::],'.',zorder=-100)\n",
    "yhat = savgol_filter(loss_list, 500, 1)\n",
    "\n",
    "\n",
    "\n",
    "#mm = np.array(loss_list[1000:math.floor(len(loss_list) / 10 ** 3) * 10 ** 3:]).reshape(-1, 100).max(axis=1)\n",
    "#print(mm)\n",
    "#max_x = np.argsort(mm, axis=0)[-10:]\n",
    "#print(max_x)\n",
    "#max_y = np.take(loss_list[1000::], max_x)\n",
    "#order = np.argsort(max_x)\n",
    "#plt.plot(max_x[order] + 1000,max_y[order],'r')\n",
    "#print(max_x)\n",
    "#print(max_y)\n",
    "\n",
    "#plt.axvline(x=1000,linestyle='--',color='green',label='100 inner epochs')\n",
    "#plt.axvline(x=17000,linestyle='--',color='purple',label='1000 inner epochs')\n",
    "#plt.axvline(x=23100,linestyle='--',color='purple',label='1st epoch')\n",
    "plt.legend(bbox_to_anchor = (1.0, 1), loc = 'upper right')\n",
    "plt.plot(x_val[1000::],yhat[1000::],'r')\n",
    "plt.grid()\n",
    "plt.show()\n",
    "\n",
    "print(np.arange(len(loss_list[1000::])) + 1000)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "dask_chunk = 5000\n",
    "\n",
    "smoothed = yhat[1000::]\n",
    "smoothed = savgol_filter(smoothed, 500, 1)\n",
    "price_series = pd.Series(smoothed)\n",
    "price_series = price_series.pct_change().to_numpy()\n",
    "plt.plot(price_series)\n",
    "plt.axhline(linestyle='--',color='red')\n",
    "plt.axvline(x=1000,linestyle='--',color='green',label='100 inner epochs')\n",
    "plt.axvline(x=17000,linestyle='--',color='purple',label='1000 inner epochs')\n",
    "plt.legend(loc = 'lower center')\n",
    "plt.title(\"Smoothed Rate of Change of Training Loss\")\n",
    "plt.xlabel(\"Training Sample (~x\" + str(int(dask_chunk)) + \")\")\n",
    "plt.ylabel(\"Rate of Change\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('med/old.picke', 'rb') as handle:\n",
    "    loss_list = pickle.load(handle)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "837"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(loss_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "42"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "loss_list = loss_list[0::20]\n",
    "len(loss_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('med/imagen_features_low_res_asis_8_24s.ptloss_total_1.picke2', 'wb') as handle:\n",
    "    pickle.dump(loss_list, handle)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "42\n"
     ]
    }
   ],
   "source": [
    "import pickle\n",
    "with open('imagen_features_low_res_asis_8_24s.ptloss_total_1.picke', 'rb') as handle:\n",
    "    loss_list = pickle.load(handle)\n",
    "    print(len(loss_list))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "import dask.dataframe as dd\n",
    "\n",
    "import dask.array as da\n",
    "\n",
    "import dask.bag as db"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><strong>Dask DataFrame Structure:</strong></div>\n",
       "<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>col1</th>\n",
       "      <th>col2</th>\n",
       "      <th>col3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>npartitions=2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>int64</td>\n",
       "      <td>int64</td>\n",
       "      <td>int64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "<div>Dask Name: from_pandas, 1 graph layer</div>"
      ],
      "text/plain": [
       "Dask DataFrame Structure:\n",
       "                col1   col2   col3\n",
       "npartitions=2                     \n",
       "0              int64  int64  int64\n",
       "2                ...    ...    ...\n",
       "3                ...    ...    ...\n",
       "Dask Name: from_pandas, 1 graph layer"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d = {'col1': [0,5,10,15], 'col2': [1,6,11,16], 'col3':[2,7,12,-1]}\n",
    "\n",
    "df = pd.DataFrame(data=d)\n",
    "\n",
    "df = dd.from_pandas(df, npartitions=2)\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "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>col1</th>\n",
       "      <th>col2</th>\n",
       "      <th>col3</th>\n",
       "      <th>image_guide</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "      <td>12</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>15</td>\n",
       "      <td>16</td>\n",
       "      <td>-1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   col1  col2  col3  image_guide\n",
       "0     0     1     2          NaN\n",
       "1     5     6     7          NaN\n",
       "2    10    11    12          NaN\n",
       "3    15    16    -1          NaN"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "df[\"image_guide\"] = np.nan\n",
    "df.compute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0,  5, 10, 15])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data5 = np.array(df['col1']).astype('int')\n",
    "data5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "metadata": {},
   "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>col1</th>\n",
       "      <th>col2</th>\n",
       "      <th>col3</th>\n",
       "      <th>image_guide</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "      <td>12</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>15</td>\n",
       "      <td>16</td>\n",
       "      <td>-1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   col1  col2  col3  image_guide\n",
       "0     0     1     2          NaN\n",
       "1     5     6     7          NaN\n",
       "2    10    11    12          NaN\n",
       "3    15    16    -1          NaN"
      ]
     },
     "execution_count": 259,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.set_index(\"col3\")\n",
    "df.repartition(npartitions=2)\n",
    "df.compute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "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>col1</th>\n",
       "      <th>col2</th>\n",
       "      <th>image_guide</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   col1  col2  image_guide\n",
       "0     1     3          NaN\n",
       "0     2     4          NaN"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.sample(frac=1).reset_index(drop=True)\n",
    "df.compute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "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>col1</th>\n",
       "      <th>col2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   col1  col2\n",
       "0     1     3\n",
       "0     2     4"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_frame = df[[\"col1\",\"col2\"]]\n",
    "data_frame.compute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "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>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [x, y, z]\n",
       "Index: []"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "meta_df = pd.DataFrame(columns=[\"x\", \"y\",\"z\"], dtype=int)\n",
    "meta_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><strong>Dask DataFrame Structure:</strong></div>\n",
       "<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>col1</th>\n",
       "      <th>col2</th>\n",
       "      <th>col3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>npartitions=2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <td>int64</td>\n",
       "      <td>int64</td>\n",
       "      <td>object</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "<div>Dask Name: assign, 8 graph layers</div>"
      ],
      "text/plain": [
       "Dask DataFrame Structure:\n",
       "                col1   col2    col3\n",
       "npartitions=2                      \n",
       "               int64  int64  object\n",
       "                 ...    ...     ...\n",
       "                 ...    ...     ...\n",
       "Dask Name: assign, 8 graph layers"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def speaker_emb_preprocess(x):\n",
    "    return x*x*x\n",
    "\n",
    "data_frame['col3'] = data_frame['col1'].apply(lambda x: speaker_emb_preprocess(x),meta=('1', 'object'))\n",
    "data_frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "str"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "meta_df = pd.DataFrame([[1,2,4],[1,4,5]], dtype=int)\n",
    "type(\"meta_df\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "list"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(np.zeros((1,768)).tolist())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><strong>Dask DataFrame Structure:</strong></div>\n",
       "<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>col2</th>\n",
       "      <th>col3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>npartitions=2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <td>int64</td>\n",
       "      <td>object</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "<div>Dask Name: drop_by_shallow_copy, 9 graph layers</div>"
      ],
      "text/plain": [
       "Dask DataFrame Structure:\n",
       "                col2    col3\n",
       "npartitions=2               \n",
       "               int64  object\n",
       "                 ...     ...\n",
       "                 ...     ...\n",
       "Dask Name: drop_by_shallow_copy, 9 graph layers"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_frame = data_frame.drop(['col1'], axis=1)\n",
    "data_frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><strong>Dask DataFrame Structure:</strong></div>\n",
       "<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>col2</th>\n",
       "      <th>col3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>npartitions=2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <td>int64</td>\n",
       "      <td>object</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "<div>Dask Name: getitem, 10 graph layers</div>"
      ],
      "text/plain": [
       "Dask DataFrame Structure:\n",
       "                col2    col3\n",
       "npartitions=2               \n",
       "               int64  object\n",
       "                 ...     ...\n",
       "                 ...     ...\n",
       "Dask Name: getitem, 10 graph layers"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_frame[['col2','col3']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([3, 4])"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(data_frame[\"col2\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n",
      "4\n"
     ]
    }
   ],
   "source": [
    "for index, row in data_frame.iterrows():\n",
    "    print(row['col2'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "'_LocIndexer' object does not support item assignment",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb Cell 14\u001b[0m line \u001b[0;36m1\n\u001b[0;32m----> <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb#X16sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m data_frame\u001b[39m.\u001b[39;49mloc[\u001b[39m1\u001b[39;49m,\u001b[39m\"\u001b[39;49m\u001b[39mcol2\u001b[39;49m\u001b[39m\"\u001b[39;49m]\u001b[39m=\u001b[39m\u001b[39m2\u001b[39m\n",
      "\u001b[0;31mTypeError\u001b[0m: '_LocIndexer' object does not support item assignment"
     ]
    }
   ],
   "source": [
    "data_frame.loc[1,\"col2\"]=2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "x = []\n",
    "x.append(0)\n",
    "x.append(1)\n",
    "x.append(2)\n",
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig = plt.figure()\n",
    "plt.plot(range(10))\n",
    "\n",
    "plt.title(\"Training Loss\")\n",
    "plt.xlabel(\"Training Sample (x500)\")\n",
    "plt.ylabel(\"MSE Loss\")\n",
    "fig.savefig(\"save_file_name.png\")\n",
    "plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "# For a single Element\n",
    "random.choices([True, False], weights=[100, 0])[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "import numpy as np\n",
    "\n",
    "\n",
    "image_size = 128\n",
    "\n",
    "def make_square(im, min_size=128, fill_color=(0, 0, 0, 0)):\n",
    "    x, y = im.size\n",
    "    size = max(min_size, x, y)\n",
    "    new_im = Image.new('RGB', (size, size), fill_color)\n",
    "    new_im.paste(im, (int((size - x) / 2), int((size - y) / 2)))\n",
    "    im.close()\n",
    "    return new_im\n",
    "\n",
    "def image_guide_preprocess_low_res_dummy(path):\n",
    "    #print(face_id)\n",
    "\n",
    "    image_guide = np.zeros((1, 768))\n",
    "    return image_guide\n",
    "\n",
    "def image_guide_preprocess_low_res(path):\n",
    "    #print(face_id)\n",
    "\n",
    "\n",
    "    if(False): #never\n",
    "        image_guide = np.zeros((49, 768))\n",
    "        #print(image_guide.shape)\n",
    "    else:\n",
    "\n",
    "        image = Image.open(path).convert('RGB')\n",
    "\n",
    "        #print(image.size)\n",
    "\n",
    "        w_s = image_size / (1+2 * 0.4)\n",
    "        h_s = image_size / (1+2 * 0.4)\n",
    "\n",
    "        image = image.crop((0.2*w_s, 0.0*h_s, 1.6*w_s, 1.4*h_s))\n",
    "        image = image.resize((128,128))\n",
    "        print(image.size)\n",
    "        #image = im\n",
    "\n",
    "        #image = image.resize((16,16))\n",
    "        im = image\n",
    "        \n",
    "\n",
    "        print('saving')\n",
    "        image.save('opop.png')\n",
    "\n",
    "        #print(np.array(image,np.float32).shape)\n",
    "\n",
    "        pix = np.array(image, np.float32)\n",
    "        pix = np.moveaxis(pix, -1, 0)\n",
    "\n",
    "        pix = pix / 255\n",
    "        image.close()\n",
    "        im.close()\n",
    "        return pix.tolist()\n",
    "    return image_guide\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "im = '/media/gamal/Passport/Datasets/VoxCeleb2Test/Voxceleb2TestFaces/mp4/id01106/45TAeIEqhcU/00001_face_1.png'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(im)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_guide_preprocess_low_res(im)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import torch\n",
    "import pandas as pd\n",
    "from skimage import io, transform\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "from torchvision import transforms, utils"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "from itertools import islice\n",
    "import csv\n",
    "import sys\n",
    "import os\n",
    "import glob\n",
    "csv.field_size_limit(sys.maxsize)\n",
    "class CsvDataset(Dataset):\n",
    "\n",
    "    def __init__(self, csv_folder):\n",
    "\n",
    "        self.csv_folder = csv_folder\n",
    "\n",
    "    def __len__(self):\n",
    "        _, _, files = next(os.walk(self.csv_folder))\n",
    "        file_count = len(files)\n",
    "        print('found ' + str(file_count) + ' files.')\n",
    "        return file_count\n",
    "\n",
    "    def __getitem__(self, index):\n",
    "        file = sorted(glob.glob(self.csv_folder + '*.csv'))[index]\n",
    "        df = pd.read_csv(file)\n",
    "        print(df.head(1))\n",
    "        df['INPUT'] = np.fromstring(s[1:-1], dtype=np.int, sep=' ')\n",
    "        #df['INPUT'] = df['INPUT'].astype('numpy.ndarray')\n",
    "        print(df['INPUT'][0])\n",
    "        torch_tensor = torch.tensor(df.head(1).values)\n",
    "        return torch_tensor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "face_dataset = CsvDataset(csv_folder='train/')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_loader = torch.utils.data.DataLoader(face_dataset,\n",
    "        batch_size=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "found 1 files.\n",
      "   Unnamed: 0                                              INPUT  \\\n",
      "0           0  [[0. 0. 0. ... 0. 0. 1.]\\n [0. 0. 0. ... 0. 0....   \n",
      "\n",
      "                                          image_path  \\\n",
      "0  [[[0.23921569 0.23921569 0.23921569 ... 0.3019...   \n",
      "\n",
      "                                       low_res_image  \n",
      "0  [[[0.24313726 0.6313726  0.6784314  0.7529412 ...  \n"
     ]
    },
    {
     "ename": "TypeError",
     "evalue": "data type 'numpy.ndarray' not understood",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb Cell 26\u001b[0m line \u001b[0;36m1\n\u001b[0;32m----> <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb#X34sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m \u001b[39mfor\u001b[39;00m x \u001b[39min\u001b[39;00m train_loader:\n\u001b[1;32m      <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb#X34sZmlsZQ%3D%3D?line=1'>2</a>\u001b[0m     \u001b[39mprint\u001b[39m(x)\n\u001b[1;32m      <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb#X34sZmlsZQ%3D%3D?line=2'>3</a>\u001b[0m     \u001b[39mbreak\u001b[39;00m\n",
      "File \u001b[0;32m~/anaconda3/envs/ds2f_m_i/lib/python3.8/site-packages/torch/utils/data/dataloader.py:633\u001b[0m, in \u001b[0;36m_BaseDataLoaderIter.__next__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    630\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_sampler_iter \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m    631\u001b[0m     \u001b[39m# TODO(https://github.com/pytorch/pytorch/issues/76750)\u001b[39;00m\n\u001b[1;32m    632\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_reset()  \u001b[39m# type: ignore[call-arg]\u001b[39;00m\n\u001b[0;32m--> 633\u001b[0m data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_next_data()\n\u001b[1;32m    634\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_num_yielded \u001b[39m+\u001b[39m\u001b[39m=\u001b[39m \u001b[39m1\u001b[39m\n\u001b[1;32m    635\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_dataset_kind \u001b[39m==\u001b[39m _DatasetKind\u001b[39m.\u001b[39mIterable \u001b[39mand\u001b[39;00m \\\n\u001b[1;32m    636\u001b[0m         \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_IterableDataset_len_called \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39mand\u001b[39;00m \\\n\u001b[1;32m    637\u001b[0m         \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_num_yielded \u001b[39m>\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_IterableDataset_len_called:\n",
      "File \u001b[0;32m~/anaconda3/envs/ds2f_m_i/lib/python3.8/site-packages/torch/utils/data/dataloader.py:677\u001b[0m, in \u001b[0;36m_SingleProcessDataLoaderIter._next_data\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    675\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_next_data\u001b[39m(\u001b[39mself\u001b[39m):\n\u001b[1;32m    676\u001b[0m     index \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_next_index()  \u001b[39m# may raise StopIteration\u001b[39;00m\n\u001b[0;32m--> 677\u001b[0m     data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_dataset_fetcher\u001b[39m.\u001b[39;49mfetch(index)  \u001b[39m# may raise StopIteration\u001b[39;00m\n\u001b[1;32m    678\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_pin_memory:\n\u001b[1;32m    679\u001b[0m         data \u001b[39m=\u001b[39m _utils\u001b[39m.\u001b[39mpin_memory\u001b[39m.\u001b[39mpin_memory(data, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_pin_memory_device)\n",
      "File \u001b[0;32m~/anaconda3/envs/ds2f_m_i/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py:51\u001b[0m, in \u001b[0;36m_MapDatasetFetcher.fetch\u001b[0;34m(self, possibly_batched_index)\u001b[0m\n\u001b[1;32m     49\u001b[0m         data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset\u001b[39m.\u001b[39m__getitems__(possibly_batched_index)\n\u001b[1;32m     50\u001b[0m     \u001b[39melse\u001b[39;00m:\n\u001b[0;32m---> 51\u001b[0m         data \u001b[39m=\u001b[39m [\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset[idx] \u001b[39mfor\u001b[39;00m idx \u001b[39min\u001b[39;00m possibly_batched_index]\n\u001b[1;32m     52\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m     53\u001b[0m     data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset[possibly_batched_index]\n",
      "File \u001b[0;32m~/anaconda3/envs/ds2f_m_i/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py:51\u001b[0m, in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m     49\u001b[0m         data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset\u001b[39m.\u001b[39m__getitems__(possibly_batched_index)\n\u001b[1;32m     50\u001b[0m     \u001b[39melse\u001b[39;00m:\n\u001b[0;32m---> 51\u001b[0m         data \u001b[39m=\u001b[39m [\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mdataset[idx] \u001b[39mfor\u001b[39;00m idx \u001b[39min\u001b[39;00m possibly_batched_index]\n\u001b[1;32m     52\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m     53\u001b[0m     data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset[possibly_batched_index]\n",
      "\u001b[1;32m/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb Cell 26\u001b[0m line \u001b[0;36m2\n\u001b[1;32m     <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb#X34sZmlsZQ%3D%3D?line=20'>21</a>\u001b[0m df \u001b[39m=\u001b[39m pd\u001b[39m.\u001b[39mread_csv(file)\n\u001b[1;32m     <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb#X34sZmlsZQ%3D%3D?line=21'>22</a>\u001b[0m \u001b[39mprint\u001b[39m(df\u001b[39m.\u001b[39mhead(\u001b[39m1\u001b[39m))\n\u001b[0;32m---> <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb#X34sZmlsZQ%3D%3D?line=22'>23</a>\u001b[0m df[\u001b[39m'\u001b[39m\u001b[39mINPUT\u001b[39m\u001b[39m'\u001b[39m] \u001b[39m=\u001b[39m df[\u001b[39m'\u001b[39;49m\u001b[39mINPUT\u001b[39;49m\u001b[39m'\u001b[39;49m]\u001b[39m.\u001b[39;49mastype(\u001b[39m'\u001b[39;49m\u001b[39mnumpy.ndarray\u001b[39;49m\u001b[39m'\u001b[39;49m)\n\u001b[1;32m     <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb#X34sZmlsZQ%3D%3D?line=23'>24</a>\u001b[0m \u001b[39m#df['INPUT'] = df['INPUT'].astype('numpy.ndarray')\u001b[39;00m\n\u001b[1;32m     <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test.ipynb#X34sZmlsZQ%3D%3D?line=24'>25</a>\u001b[0m \u001b[39mprint\u001b[39m(\u001b[39mtype\u001b[39m(df[\u001b[39m'\u001b[39m\u001b[39mINPUT\u001b[39m\u001b[39m'\u001b[39m]\u001b[39m.\u001b[39mdtypes))\n",
      "File \u001b[0;32m~/anaconda3/envs/ds2f_m_i/lib/python3.8/site-packages/pandas/core/generic.py:6240\u001b[0m, in \u001b[0;36mNDFrame.astype\u001b[0;34m(self, dtype, copy, errors)\u001b[0m\n\u001b[1;32m   6233\u001b[0m     results \u001b[39m=\u001b[39m [\n\u001b[1;32m   6234\u001b[0m         \u001b[39mself\u001b[39m\u001b[39m.\u001b[39miloc[:, i]\u001b[39m.\u001b[39mastype(dtype, copy\u001b[39m=\u001b[39mcopy)\n\u001b[1;32m   6235\u001b[0m         \u001b[39mfor\u001b[39;00m i \u001b[39min\u001b[39;00m \u001b[39mrange\u001b[39m(\u001b[39mlen\u001b[39m(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcolumns))\n\u001b[1;32m   6236\u001b[0m     ]\n\u001b[1;32m   6238\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m   6239\u001b[0m     \u001b[39m# else, only a single dtype is given\u001b[39;00m\n\u001b[0;32m-> 6240\u001b[0m     new_data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_mgr\u001b[39m.\u001b[39;49mastype(dtype\u001b[39m=\u001b[39;49mdtype, copy\u001b[39m=\u001b[39;49mcopy, errors\u001b[39m=\u001b[39;49merrors)\n\u001b[1;32m   6241\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_constructor(new_data)\u001b[39m.\u001b[39m__finalize__(\u001b[39mself\u001b[39m, method\u001b[39m=\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mastype\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[1;32m   6243\u001b[0m \u001b[39m# GH 33113: handle empty frame or series\u001b[39;00m\n",
      "File \u001b[0;32m~/anaconda3/envs/ds2f_m_i/lib/python3.8/site-packages/pandas/core/internals/managers.py:448\u001b[0m, in \u001b[0;36mBaseBlockManager.astype\u001b[0;34m(self, dtype, copy, errors)\u001b[0m\n\u001b[1;32m    447\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mastype\u001b[39m(\u001b[39mself\u001b[39m: T, dtype, copy: \u001b[39mbool\u001b[39m \u001b[39m=\u001b[39m \u001b[39mFalse\u001b[39;00m, errors: \u001b[39mstr\u001b[39m \u001b[39m=\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mraise\u001b[39m\u001b[39m\"\u001b[39m) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m T:\n\u001b[0;32m--> 448\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mapply(\u001b[39m\"\u001b[39;49m\u001b[39mastype\u001b[39;49m\u001b[39m\"\u001b[39;49m, dtype\u001b[39m=\u001b[39;49mdtype, copy\u001b[39m=\u001b[39;49mcopy, errors\u001b[39m=\u001b[39;49merrors)\n",
      "File \u001b[0;32m~/anaconda3/envs/ds2f_m_i/lib/python3.8/site-packages/pandas/core/internals/managers.py:352\u001b[0m, in \u001b[0;36mBaseBlockManager.apply\u001b[0;34m(self, f, align_keys, ignore_failures, **kwargs)\u001b[0m\n\u001b[1;32m    350\u001b[0m         applied \u001b[39m=\u001b[39m b\u001b[39m.\u001b[39mapply(f, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\n\u001b[1;32m    351\u001b[0m     \u001b[39melse\u001b[39;00m:\n\u001b[0;32m--> 352\u001b[0m         applied \u001b[39m=\u001b[39m \u001b[39mgetattr\u001b[39;49m(b, f)(\u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m    353\u001b[0m \u001b[39mexcept\u001b[39;00m (\u001b[39mTypeError\u001b[39;00m, \u001b[39mNotImplementedError\u001b[39;00m):\n\u001b[1;32m    354\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m ignore_failures:\n",
      "File \u001b[0;32m~/anaconda3/envs/ds2f_m_i/lib/python3.8/site-packages/pandas/core/internals/blocks.py:526\u001b[0m, in \u001b[0;36mBlock.astype\u001b[0;34m(self, dtype, copy, errors)\u001b[0m\n\u001b[1;32m    508\u001b[0m \u001b[39m\u001b[39m\u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m    509\u001b[0m \u001b[39mCoerce to the new dtype.\u001b[39;00m\n\u001b[1;32m    510\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    522\u001b[0m \u001b[39mBlock\u001b[39;00m\n\u001b[1;32m    523\u001b[0m \u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m    524\u001b[0m values \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mvalues\n\u001b[0;32m--> 526\u001b[0m new_values \u001b[39m=\u001b[39m astype_array_safe(values, dtype, copy\u001b[39m=\u001b[39;49mcopy, errors\u001b[39m=\u001b[39;49merrors)\n\u001b[1;32m    528\u001b[0m new_values \u001b[39m=\u001b[39m maybe_coerce_values(new_values)\n\u001b[1;32m    529\u001b[0m newb \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mmake_block(new_values)\n",
      "File \u001b[0;32m~/anaconda3/envs/ds2f_m_i/lib/python3.8/site-packages/pandas/core/dtypes/astype.py:279\u001b[0m, in \u001b[0;36mastype_array_safe\u001b[0;34m(values, dtype, copy, errors)\u001b[0m\n\u001b[1;32m    273\u001b[0m     msg \u001b[39m=\u001b[39m (\n\u001b[1;32m    274\u001b[0m         \u001b[39mf\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mExpected an instance of \u001b[39m\u001b[39m{\u001b[39;00mdtype\u001b[39m.\u001b[39m\u001b[39m__name__\u001b[39m\u001b[39m}\u001b[39;00m\u001b[39m, \u001b[39m\u001b[39m\"\u001b[39m\n\u001b[1;32m    275\u001b[0m         \u001b[39m\"\u001b[39m\u001b[39mbut got the class instead. Try instantiating \u001b[39m\u001b[39m'\u001b[39m\u001b[39mdtype\u001b[39m\u001b[39m'\u001b[39m\u001b[39m.\u001b[39m\u001b[39m\"\u001b[39m\n\u001b[1;32m    276\u001b[0m     )\n\u001b[1;32m    277\u001b[0m     \u001b[39mraise\u001b[39;00m \u001b[39mTypeError\u001b[39;00m(msg)\n\u001b[0;32m--> 279\u001b[0m dtype \u001b[39m=\u001b[39m pandas_dtype(dtype)\n\u001b[1;32m    280\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(dtype, PandasDtype):\n\u001b[1;32m    281\u001b[0m     \u001b[39m# Ensure we don't end up with a PandasArray\u001b[39;00m\n\u001b[1;32m    282\u001b[0m     dtype \u001b[39m=\u001b[39m dtype\u001b[39m.\u001b[39mnumpy_dtype\n",
      "File \u001b[0;32m~/anaconda3/envs/ds2f_m_i/lib/python3.8/site-packages/pandas/core/dtypes/common.py:1781\u001b[0m, in \u001b[0;36mpandas_dtype\u001b[0;34m(dtype)\u001b[0m\n\u001b[1;32m   1778\u001b[0m \u001b[39m# try a numpy dtype\u001b[39;00m\n\u001b[1;32m   1779\u001b[0m \u001b[39m# raise a consistent TypeError if failed\u001b[39;00m\n\u001b[1;32m   1780\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m-> 1781\u001b[0m     npdtype \u001b[39m=\u001b[39m np\u001b[39m.\u001b[39;49mdtype(dtype)\n\u001b[1;32m   1782\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mSyntaxError\u001b[39;00m \u001b[39mas\u001b[39;00m err:\n\u001b[1;32m   1783\u001b[0m     \u001b[39m# np.dtype uses `eval` which can raise SyntaxError\u001b[39;00m\n\u001b[1;32m   1784\u001b[0m     \u001b[39mraise\u001b[39;00m \u001b[39mTypeError\u001b[39;00m(\u001b[39mf\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mdata type \u001b[39m\u001b[39m'\u001b[39m\u001b[39m{\u001b[39;00mdtype\u001b[39m}\u001b[39;00m\u001b[39m'\u001b[39m\u001b[39m not understood\u001b[39m\u001b[39m\"\u001b[39m) \u001b[39mfrom\u001b[39;00m \u001b[39merr\u001b[39;00m\n",
      "\u001b[0;31mTypeError\u001b[0m: data type 'numpy.ndarray' not understood"
     ]
    }
   ],
   "source": [
    "for x in train_loader:\n",
    "    print(x)\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "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>(i0, i1, i2, i3, i4, i5, i6, i7, i8, i9)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [(i0, i1, i2, i3, i4, i5, i6, i7, i8, i9)]\n",
       "Index: []"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "names = np.arange(10)\n",
    "charar = np.chararray((1, 10))\n",
    "charar[:] = 'i'\n",
    "names = charar.astype('str') + names.astype('str')\n",
    "meta_df = pd.DataFrame(columns=names.astype(str), dtype=int)\n",
    "meta_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['0', '1', '2', '3', '4', '5', '6', '7', '8', '9'], dtype='<U21')"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'np' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m/home/gamal/vsc/DiffusionSpeech2Face/test copy.ipynb Cell 40\u001b[0m line \u001b[0;36m<cell line: 3>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test%20copy.ipynb#X54sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m lr_lambda \u001b[39m=\u001b[39m \u001b[39mlambda\u001b[39;00m step: \u001b[39m1.0\u001b[39m\n\u001b[0;32m----> <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test%20copy.ipynb#X54sZmlsZQ%3D%3D?line=2'>3</a>\u001b[0m a \u001b[39m=\u001b[39m np\u001b[39m.\u001b[39maran\n\u001b[1;32m      <a href='vscode-notebook-cell:/home/gamal/vsc/DiffusionSpeech2Face/test%20copy.ipynb#X54sZmlsZQ%3D%3D?line=3'>4</a>\u001b[0m lr_lambda(\u001b[39m3\u001b[39m)\n",
      "\u001b[0;31mNameError\u001b[0m: name 'np' is not defined"
     ]
    }
   ],
   "source": [
    "lr_lambda = lambda step: 1.0\n",
    "\n",
    "a = np.aran\n",
    "lr_lambda(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "fff = [[1,2],[2,3]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[2, 3, 22]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_val = [x[1] for x in fff]\n",
    "x_val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[1, 2], [2, 3], [11, 22]]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fff.append([11,22])\n",
    "fff"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "UNET = 1\n",
    "model_filename = '/media/gamal/Passport/checkpoint/noablation'\n",
    "from pathlib import Path\n",
    "import pickle\n",
    "from scipy.signal import savgol_filter\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "if(UNET == 1):\n",
    "    my_file = Path(model_filename + '/loss_total_1.picke')\n",
    "    if my_file.is_file():\n",
    "        with open(model_filename + '/loss_total_1.picke', 'rb') as handle:\n",
    "            loss_total = pickle.load(handle)\n",
    "    else:\n",
    "        loss_total = []\n",
    "else:\n",
    "    my_file = Path(model_filename + '/loss_total_2.picke')\n",
    "    if my_file.is_file():\n",
    "        with open(model_filename + '/loss_total_2.picke', 'rb') as handle:\n",
    "            loss_total = pickle.load(handle)\n",
    "    else:\n",
    "        loss_total = []\n",
    "\n",
    "#print(loss_list)\n",
    "#print(loss_total)\n",
    "\n",
    "\n",
    "#print(loss_total)\n",
    "x_val = [x[0] for x in loss_total]\n",
    "y_val = [x[1] for x in loss_total]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.8911876194179058,\n",
       " 0.7999346889555454,\n",
       " 0.8537470139563084,\n",
       " 0.7711716834455729,\n",
       " 0.8232662603259087,\n",
       " 0.7347892560064793,\n",
       " 0.7516805175691843,\n",
       " 0.7773085813969374,\n",
       " 0.7015355862677097,\n",
       " 0.6758643612265587,\n",
       " 0.6007436234503984,\n",
       " 0.6061839666217566,\n",
       " 0.4894444076344371,\n",
       " 0.42927634343504906,\n",
       " 0.4027000442147255,\n",
       " 0.36129788495600224,\n",
       " 0.3116556415334344,\n",
       " 0.26066588424146175,\n",
       " 0.28484330140054226,\n",
       " 0.37355291470885277,\n",
       " 0.22235299739986658,\n",
       " 0.26791768334805965,\n",
       " 0.2272719694301486,\n",
       " 0.2319795759394765,\n",
       " 0.18687127763405442,\n",
       " 0.19665296701714396,\n",
       " 0.1882832869887352,\n",
       " 0.17905147140845656,\n",
       " 0.16436071507632732,\n",
       " 0.15766430227085948,\n",
       " 0.1517795091494918,\n",
       " 0.1472784816287458,\n",
       " 0.1341643687337637,\n",
       " 0.13444533525034785,\n",
       " 0.14589114440605044,\n",
       " 0.13988698413595557,\n",
       " 0.14999722316861153,\n",
       " 0.12589075695723295,\n",
       " 0.1194712808355689,\n",
       " 0.12022729963064194,\n",
       " 0.11880913423374295,\n",
       " 0.11935455817729235,\n",
       " 0.11919081537052989,\n",
       " 0.11077925702556968,\n",
       " 0.11163061577826738,\n",
       " 0.11463105445727706,\n",
       " 0.12227463722229004,\n",
       " 0.10859032231383026,\n",
       " 0.10553807625547051,\n",
       " 0.10176883498206735,\n",
       " 0.11115950183011591,\n",
       " 0.10437851678580046,\n",
       " 0.09973209467716515,\n",
       " 0.10652324697002769,\n",
       " 0.08982553146779537,\n",
       " 0.08770293951965868,\n",
       " 0.09421770554035902,\n",
       " 0.08892158744856715,\n",
       " 0.09052566438913345,\n",
       " 0.09084778372198343,\n",
       " 0.09024817636236548,\n",
       " 0.09556482452899218,\n",
       " 0.09640506468713284,\n",
       " 0.0934225576929748,\n",
       " 0.09334817714989185,\n",
       " 0.08524103881791234,\n",
       " 0.08772248192690313,\n",
       " 0.09167913580313325,\n",
       " 0.08374922117218375,\n",
       " 0.0937956883572042,\n",
       " 0.09093782166019082,\n",
       " 0.08454366796649992,\n",
       " 0.08597583649680018,\n",
       " 0.08051008265465498,\n",
       " 0.0808218342717737,\n",
       " 0.08269596192985773,\n",
       " 0.08998717297799885,\n",
       " 0.08024744712747633,\n",
       " 0.08092610072344542,\n",
       " 0.07500219508074224,\n",
       " 0.08169546234421432,\n",
       " 0.08363931765779853,\n",
       " 0.07889737910591066,\n",
       " 0.08487832779064775,\n",
       " 0.06763727613724768,\n",
       " 0.07689591008238494,\n",
       " 0.07752389856614172,\n",
       " 0.07057361514307559,\n",
       " 0.0730672744102776,\n",
       " 0.07279044738970697,\n",
       " 0.07264092238619924,\n",
       " 0.08019614545628428,\n",
       " 0.060706437565386295,\n",
       " 0.06676208763383329,\n",
       " 0.06877836119383574,\n",
       " 0.06826721481047571,\n",
       " 0.06957211392000318,\n",
       " 0.06693843891844153,\n",
       " 0.06672955537214875,\n",
       " 0.06446325173601508,\n",
       " 0.0725670384708792,\n",
       " 0.06172327185049653,\n",
       " 0.06151972874067724,\n",
       " 0.06052645551972091,\n",
       " 0.06214160588569939,\n",
       " 0.05507611948996782,\n",
       " 0.057316195452585816,\n",
       " 0.06973568419925869,\n",
       " 0.0611052515450865,\n",
       " 0.060711396392434835,\n",
       " 0.05984937003813684,\n",
       " 0.05657303845509887,\n",
       " 0.0562876935582608,\n",
       " 0.060411642654798925,\n",
       " 0.06942980038002133,\n",
       " 0.06043257797136903,\n",
       " 0.06008081906475127,\n",
       " 0.05827025533653796,\n",
       " 0.06317000067792833,\n",
       " 0.058706947369500995,\n",
       " 0.05678856582380831,\n",
       " 0.052494254894554615,\n",
       " 0.050112121272832155,\n",
       " 0.053212708327919245,\n",
       " 0.055576415499672294,\n",
       " 0.04681564727798104,\n",
       " 0.05402604315895587,\n",
       " 0.058267006068490446,\n",
       " 0.04752727167215198,\n",
       " 0.04843936744146049,\n",
       " 0.045388598926365376,\n",
       " 0.04676171066239476,\n",
       " 0.047411266481503844,\n",
       " 0.05211783782579005,\n",
       " 0.049143296550028026,\n",
       " 0.05424246028997004,\n",
       " 0.05222422175575048,\n",
       " 0.0516881353687495,\n",
       " 0.047535783844068646,\n",
       " 0.04775120201520622,\n",
       " 0.0487530967220664,\n",
       " 0.04465352580882609,\n",
       " 0.045695146545767784]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1129,\n",
       " 1137,\n",
       " 1142,\n",
       " 1156,\n",
       " 1165,\n",
       " 2031,\n",
       " 2040,\n",
       " 2049,\n",
       " 2063,\n",
       " 2074,\n",
       " 2084,\n",
       " 2095,\n",
       " 2103,\n",
       " 2115,\n",
       " 2125,\n",
       " 2136,\n",
       " 2148,\n",
       " 2159,\n",
       " 2169,\n",
       " 2178,\n",
       " 2192,\n",
       " 3054,\n",
       " 3066,\n",
       " 3077,\n",
       " 3085,\n",
       " 3098,\n",
       " 3112,\n",
       " 3116,\n",
       " 3127,\n",
       " 3135,\n",
       " 3143,\n",
       " 3158,\n",
       " 3172,\n",
       " 3178,\n",
       " 3188,\n",
       " 3200,\n",
       " 3211,\n",
       " 4082,\n",
       " 4095,\n",
       " 4108,\n",
       " 4119,\n",
       " 4126,\n",
       " 4138,\n",
       " 4151,\n",
       " 4163,\n",
       " 4167,\n",
       " 4181,\n",
       " 4193,\n",
       " 4204,\n",
       " 4210,\n",
       " 4223,\n",
       " 4236,\n",
       " 4248,\n",
       " 5104,\n",
       " 5118,\n",
       " 5122,\n",
       " 5134,\n",
       " 5148,\n",
       " 5160,\n",
       " 5163,\n",
       " 5176,\n",
       " 5188,\n",
       " 5202,\n",
       " 5205,\n",
       " 5218,\n",
       " 5224,\n",
       " 5234,\n",
       " 5248,\n",
       " 5260,\n",
       " 6129,\n",
       " 6140,\n",
       " 6148,\n",
       " 6158,\n",
       " 6172,\n",
       " 6183,\n",
       " 6191,\n",
       " 6202,\n",
       " 6212,\n",
       " 6224,\n",
       " 6229,\n",
       " 6240,\n",
       " 6254,\n",
       " 6267,\n",
       " 6278,\n",
       " 6285,\n",
       " 7155,\n",
       " 7166,\n",
       " 7175,\n",
       " 7185,\n",
       " 7197,\n",
       " 7211,\n",
       " 7222,\n",
       " 7231,\n",
       " 7242,\n",
       " 7255,\n",
       " 7258,\n",
       " 7269,\n",
       " 7283,\n",
       " 7296,\n",
       " 7301,\n",
       " 7313,\n",
       " 8180,\n",
       " 8188,\n",
       " 8196,\n",
       " 8206,\n",
       " 8217,\n",
       " 8231,\n",
       " 8240,\n",
       " 8252,\n",
       " 8258,\n",
       " 8270,\n",
       " 8281,\n",
       " 8293,\n",
       " 8304,\n",
       " 8313,\n",
       " 8324,\n",
       " 8338,\n",
       " 9198,\n",
       " 9210,\n",
       " 9221,\n",
       " 9232,\n",
       " 9238,\n",
       " 9250,\n",
       " 9259,\n",
       " 9270,\n",
       " 9279,\n",
       " 9286,\n",
       " 9298,\n",
       " 9307,\n",
       " 9319,\n",
       " 9332,\n",
       " 9340,\n",
       " 9349]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_val[10::]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "ename": "SyntaxError",
     "evalue": "invalid syntax (318621653.py, line 10)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;36m  Input \u001b[0;32mIn [27]\u001b[0;36m\u001b[0m\n\u001b[0;31m    yhat = savgol_filter(y_np.arange(len(loss_total[10::])) + 10val, 100, 3)\u001b[0m\n\u001b[0m                                                                ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m invalid syntax\n"
     ]
    }
   ],
   "source": [
    "\n",
    "fig = plt.figure()\n",
    "plt.plot(x_val[10::],np.arange(len(y_val[10::])) + 10,y_val[10::]\n",
    "        ,'.',zorder=-100)\n",
    "\n",
    "#plt.axvline(x=1000,linestyle='--',color='green',label='1000 inner epochs')\n",
    "#plt.axvline(x=17000,linestyle='--',color='purple',label='100 inner epochs - 100000 unique samples seen' )\n",
    "#plt.axvline(x=23100,linestyle='-.',color='black',label='1 inner epoch - end of 1st epoch ')\n",
    "plt.legend(bbox_to_anchor = (1.0, 1), loc = 'upper right')\n",
    "plt.grid()\n",
    "yhat = savgol_filter(y_val, 100, 3)\n",
    "#plt.plot(x_val,yhat[10::],'r')\n",
    "plt.title(\"Training Loss\")\n",
    "plt.xlabel(\"Training Sample\")\n",
    "plt.ylabel(\"MSE Loss\")\n",
    "\n",
    "\n",
    "\n",
    "plt.show()\n",
    "\n",
    "\n",
    "plt.close()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ds2f_pre",
   "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": "3.8.16"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}