bio_ML_2019.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e069f78b-33be-4216-9db6-ff20138a5886",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/phucdt/anaconda3/envs/bio/lib/python3.9/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "from numpy import reshape\n",
    "import seaborn as sns\n",
    "import pandas as pd  \n",
    "import torch\n",
    "from os import listdir\n",
    "from os.path import isfile, join\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a639560f-a181-49a7-b4f8-ae7709957e99",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50224/50224 [03:41<00:00, 227.10it/s]\n"
     ]
    }
   ],
   "source": [
    "mypath = \"./asvspoof2021/LA/Baseline-RawNet2-bio/feats/train_2019/\"\n",
    "onlyfiles = [f for f in listdir(mypath) if isfile(join(mypath, f))]\n",
    "feat = pd.DataFrame()\n",
    "# feat.columns = [\"utt\", \"feats\"]\n",
    "\n",
    "for f in tqdm(onlyfiles):\n",
    "    # print(f)\n",
    "    bio_inp = torch.load(join(mypath, f)).cpu().detach().numpy()\n",
    "    feat = feat.append({\"utt\": f, \"feats\": bio_inp}, ignore_index=True)\n",
    "    # print(bio_inp)\n",
    "    \n",
    "# feat.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "0b24e64c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 25379 entries, 0 to 25378\n",
      "Data columns (total 6 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   sid     25379 non-null  object\n",
      " 1   utt     25379 non-null  object\n",
      " 2   phy     25379 non-null  object\n",
      " 3   attack  25379 non-null  object\n",
      " 4   class   25379 non-null  object\n",
      " 5   feats   25379 non-null  object\n",
      "dtypes: object(6)\n",
      "memory usage: 1.2+ MB\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 17779 entries, 0 to 17778\n",
      "Data columns (total 6 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   sid     17779 non-null  object\n",
      " 1   utt     17779 non-null  object\n",
      " 2   phy     17779 non-null  object\n",
      " 3   attack  17779 non-null  object\n",
      " 4   class   17779 non-null  object\n",
      " 5   feats   17779 non-null  object\n",
      "dtypes: object(6)\n",
      "memory usage: 972.3+ KB\n"
     ]
    }
   ],
   "source": [
    "df = pd.read_csv('/dataa/Dataset/ASVspoof/LA/ASVspoof2019_LA_cm_protocols/ASVspoof2019.LA.cm.train.trn.txt', sep=' ')\n",
    "df.columns = [\"sid\", \"utt\", \"phy\", \"attack\", \"class\"]\n",
    "full = df.join(feat.set_index('utt'), on=\"utt\")\n",
    "full.info()\n",
    "tts_full_train = full.loc[~full[\"attack\"].str.contains(\"A05\") & ~full[\"attack\"].str.contains(\"A06\")]\n",
    "tts_full_train.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "16992053-69ee-4520-8aef-45d40f424c8c",
   "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>sid</th>\n",
       "      <th>utt</th>\n",
       "      <th>phy</th>\n",
       "      <th>attack</th>\n",
       "      <th>class</th>\n",
       "      <th>feats</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LA_0079</td>\n",
       "      <td>LA_T_1271820</td>\n",
       "      <td>-</td>\n",
       "      <td>-</td>\n",
       "      <td>bonafide</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>LA_0079</td>\n",
       "      <td>LA_T_1272637</td>\n",
       "      <td>-</td>\n",
       "      <td>-</td>\n",
       "      <td>bonafide</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LA_0079</td>\n",
       "      <td>LA_T_1276960</td>\n",
       "      <td>-</td>\n",
       "      <td>-</td>\n",
       "      <td>bonafide</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LA_0079</td>\n",
       "      <td>LA_T_1341447</td>\n",
       "      <td>-</td>\n",
       "      <td>-</td>\n",
       "      <td>bonafide</td>\n",
       "      <td>[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LA_0079</td>\n",
       "      <td>LA_T_1363611</td>\n",
       "      <td>-</td>\n",
       "      <td>-</td>\n",
       "      <td>bonafide</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       sid           utt phy attack     class  \\\n",
       "0  LA_0079  LA_T_1271820   -      -  bonafide   \n",
       "1  LA_0079  LA_T_1272637   -      -  bonafide   \n",
       "2  LA_0079  LA_T_1276960   -      -  bonafide   \n",
       "3  LA_0079  LA_T_1341447   -      -  bonafide   \n",
       "4  LA_0079  LA_T_1363611   -      -  bonafide   \n",
       "\n",
       "                                               feats  \n",
       "0  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "1  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "2  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "3  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...  \n",
       "4  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full = df.join(feat.set_index('utt'), on=\"utt\")\n",
    "tts_full_train = full.loc[~full[\"attack\"].str.contains(\"A05\") & ~full[\"attack\"].str.contains(\"A06\")]\n",
    "tts_full_train.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6599ab97-9ca7-49a2-85c4-4cd06023412e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 25379 entries, 0 to 25378\n",
      "Data columns (total 6 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   sid     25379 non-null  object\n",
      " 1   utt     25379 non-null  object\n",
      " 2   phy     25379 non-null  object\n",
      " 3   attack  25379 non-null  object\n",
      " 4   class   25379 non-null  object\n",
      " 5   feats   25379 non-null  object\n",
      "dtypes: object(6)\n",
      "memory usage: 1.2+ MB\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 17779 entries, 0 to 17778\n",
      "Data columns (total 6 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   sid     17779 non-null  object\n",
      " 1   utt     17779 non-null  object\n",
      " 2   phy     17779 non-null  object\n",
      " 3   attack  17779 non-null  object\n",
      " 4   class   17779 non-null  object\n",
      " 5   feats   17779 non-null  object\n",
      "dtypes: object(6)\n",
      "memory usage: 972.3+ KB\n"
     ]
    }
   ],
   "source": [
    "df = pd.read_csv('/root/dataset/ASVspoof/LA/ASVspoof2019_LA_cm_protocols/ASVspoof2019.LA.cm.train.trn.txt', sep=' ')\n",
    "df.columns = [\"sid\", \"utt\", \"phy\", \"attack\", \"class\"]\n",
    "full = df.join(feat.set_index('utt'), on=\"utt\")\n",
    "full.info()\n",
    "tts_full_train = full.loc[~full[\"attack\"].str.contains(\"A05\") & ~full[\"attack\"].str.contains(\"A06\")]\n",
    "tts_full_train.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0c5b3ff2-7c40-4537-b521-8641f4fccb55",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 24843 entries, 0 to 24842\n",
      "Data columns (total 6 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   sid     24843 non-null  object\n",
      " 1   utt     24843 non-null  object\n",
      " 2   phy     24843 non-null  object\n",
      " 3   attack  24843 non-null  object\n",
      " 4   class   24843 non-null  object\n",
      " 5   feats   24843 non-null  object\n",
      "dtypes: object(6)\n",
      "memory usage: 1.1+ MB\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 17411 entries, 0 to 17410\n",
      "Data columns (total 6 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   sid     17411 non-null  object\n",
      " 1   utt     17411 non-null  object\n",
      " 2   phy     17411 non-null  object\n",
      " 3   attack  17411 non-null  object\n",
      " 4   class   17411 non-null  object\n",
      " 5   feats   17411 non-null  object\n",
      "dtypes: object(6)\n",
      "memory usage: 952.2+ KB\n"
     ]
    }
   ],
   "source": [
    "df = pd.read_csv('/root/dataset/ASVspoof/LA/ASVspoof2019_LA_cm_protocols/ASVspoof2019.LA.cm.dev.trl.txt', sep=' ')\n",
    "df.columns = [\"sid\", \"utt\", \"phy\", \"attack\", \"class\"]\n",
    "full = df.join(feat.set_index('utt'), on=\"utt\")\n",
    "full.info()\n",
    "tts_full_dev = full.loc[~full[\"attack\"].str.contains(\"A05\") & ~full[\"attack\"].str.contains(\"A06\")]\n",
    "tts_full_dev.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f7f9d9d4-1e63-4e1f-9cd8-9304ee19c8ba",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'tts_full_dev' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[7], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m tts_full_dev\u001b[39m.\u001b[39mhead()\n",
      "\u001b[0;31mNameError\u001b[0m: name 'tts_full_dev' is not defined"
     ]
    }
   ],
   "source": [
    "tts_full_dev.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "256a25a4-e566-40c1-957a-d4da980150c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tsnecuda import TSNE\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "82bfb20d-3dae-4ce1-8d76-ec43359128f8",
   "metadata": {},
   "outputs": [
    {
     "ename": "OSError",
     "evalue": "/opt/intel/mkl/lib/intel64/libmkl_sequential.so: undefined symbol: mkl_sparse_z_bsr_ng_avx521_sp_mv_i4",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mOSError\u001b[0m                                   Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[12], line 5\u001b[0m\n\u001b[1;32m      2\u001b[0m y \u001b[39m=\u001b[39m tts_full_train[\u001b[39m\"\u001b[39m\u001b[39mclass\u001b[39m\u001b[39m\"\u001b[39m]\u001b[39m.\u001b[39mto_list()\n\u001b[1;32m      3\u001b[0m y \u001b[39m=\u001b[39m tts_full_train[\u001b[39m\"\u001b[39m\u001b[39mattack\u001b[39m\u001b[39m\"\u001b[39m]\u001b[39m.\u001b[39mreplace(\u001b[39m\"\u001b[39m\u001b[39mA05\u001b[39m\u001b[39m\"\u001b[39m, \u001b[39m\"\u001b[39m\u001b[39mVC\u001b[39m\u001b[39m\"\u001b[39m)\u001b[39m.\u001b[39mreplace(\u001b[39m\"\u001b[39m\u001b[39mA06\u001b[39m\u001b[39m\"\u001b[39m, \u001b[39m\"\u001b[39m\u001b[39mVC\u001b[39m\u001b[39m\"\u001b[39m)\u001b[39m.\u001b[39mreplace(\u001b[39m\"\u001b[39m\u001b[39mA01\u001b[39m\u001b[39m\"\u001b[39m, \u001b[39m\"\u001b[39m\u001b[39mTTS\u001b[39m\u001b[39m\"\u001b[39m)\u001b[39m.\u001b[39mreplace(\u001b[39m\"\u001b[39m\u001b[39mA02\u001b[39m\u001b[39m\"\u001b[39m, \u001b[39m\"\u001b[39m\u001b[39mTTS\u001b[39m\u001b[39m\"\u001b[39m)\u001b[39m.\u001b[39mreplace(\u001b[39m\"\u001b[39m\u001b[39mA03\u001b[39m\u001b[39m\"\u001b[39m, \u001b[39m\"\u001b[39m\u001b[39mTTS\u001b[39m\u001b[39m\"\u001b[39m)\u001b[39m.\u001b[39mreplace(\u001b[39m\"\u001b[39m\u001b[39mA04\u001b[39m\u001b[39m\"\u001b[39m, \u001b[39m\"\u001b[39m\u001b[39mTTS\u001b[39m\u001b[39m\"\u001b[39m)\u001b[39m.\u001b[39mreplace(\u001b[39m\"\u001b[39m\u001b[39m-\u001b[39m\u001b[39m\"\u001b[39m,\u001b[39m\"\u001b[39m\u001b[39mbonafide\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[0;32m----> 5\u001b[0m tsne \u001b[39m=\u001b[39m TSNE(n_components\u001b[39m=\u001b[39;49m\u001b[39m2\u001b[39;49m, verbose\u001b[39m=\u001b[39;49m\u001b[39m1\u001b[39;49m, perplexity\u001b[39m=\u001b[39;49m\u001b[39m15\u001b[39;49m, learning_rate\u001b[39m=\u001b[39;49m\u001b[39m10\u001b[39;49m)\n\u001b[1;32m      6\u001b[0m \u001b[39m# tsne = TSNE(n_components=2, perplexity=15, learning_rate=10)\u001b[39;00m\n\u001b[1;32m      7\u001b[0m z \u001b[39m=\u001b[39m tsne\u001b[39m.\u001b[39mfit_transform(x)\n",
      "File \u001b[0;32m~/anaconda3/envs/bio/lib/python3.9/site-packages/tsnecuda/TSNE.py:139\u001b[0m, in \u001b[0;36mTSNE.__init__\u001b[0;34m(self, n_components, perplexity, early_exaggeration, learning_rate, num_neighbors, force_magnify_iters, pre_momentum, post_momentum, theta, epssq, n_iter, n_iter_without_progress, min_grad_norm, perplexity_epsilon, metric, init, return_style, num_snapshots, verbose, random_seed, use_interactive, viz_timeout, viz_server, dump_points, dump_file, dump_interval, print_interval, device, magnitude_factor)\u001b[0m\n\u001b[1;32m    137\u001b[0m \u001b[39m# Build the hooks for the BH T-SNE library\u001b[39;00m\n\u001b[1;32m    138\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_path \u001b[39m=\u001b[39m os\u001b[39m.\u001b[39mpath\u001b[39m.\u001b[39mdirname(\u001b[39m__file__\u001b[39m)\n\u001b[0;32m--> 139\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_lib \u001b[39m=\u001b[39m N\u001b[39m.\u001b[39;49mctypeslib\u001b[39m.\u001b[39;49mload_library(\n\u001b[1;32m    140\u001b[0m     \u001b[39m'\u001b[39;49m\u001b[39mlibtsnecuda\u001b[39;49m\u001b[39m'\u001b[39;49m, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_path)  \u001b[39m# Load the ctypes library\u001b[39;00m\n\u001b[1;32m    142\u001b[0m \u001b[39m# Hook the BH T-SNE function\u001b[39;00m\n\u001b[1;32m    143\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_lib\u001b[39m.\u001b[39mpymodule_tsne\u001b[39m.\u001b[39mrestype \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m\n",
      "File \u001b[0;32m~/anaconda3/envs/bio/lib/python3.9/site-packages/numpy/ctypeslib.py:149\u001b[0m, in \u001b[0;36mload_library\u001b[0;34m(libname, loader_path)\u001b[0m\n\u001b[1;32m    147\u001b[0m \u001b[39mif\u001b[39;00m os\u001b[39m.\u001b[39mpath\u001b[39m.\u001b[39mexists(libpath):\n\u001b[1;32m    148\u001b[0m     \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m--> 149\u001b[0m         \u001b[39mreturn\u001b[39;00m ctypes\u001b[39m.\u001b[39;49mcdll[libpath]\n\u001b[1;32m    150\u001b[0m     \u001b[39mexcept\u001b[39;00m \u001b[39mOSError\u001b[39;00m:\n\u001b[1;32m    151\u001b[0m         \u001b[39m## defective lib file\u001b[39;00m\n\u001b[1;32m    152\u001b[0m         \u001b[39mraise\u001b[39;00m\n",
      "File \u001b[0;32m~/anaconda3/envs/bio/lib/python3.9/ctypes/__init__.py:457\u001b[0m, in \u001b[0;36mLibraryLoader.__getitem__\u001b[0;34m(self, name)\u001b[0m\n\u001b[1;32m    456\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__getitem__\u001b[39m(\u001b[39mself\u001b[39m, name):\n\u001b[0;32m--> 457\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mgetattr\u001b[39;49m(\u001b[39mself\u001b[39;49m, name)\n",
      "File \u001b[0;32m~/anaconda3/envs/bio/lib/python3.9/ctypes/__init__.py:452\u001b[0m, in \u001b[0;36mLibraryLoader.__getattr__\u001b[0;34m(self, name)\u001b[0m\n\u001b[1;32m    450\u001b[0m \u001b[39mif\u001b[39;00m name[\u001b[39m0\u001b[39m] \u001b[39m==\u001b[39m \u001b[39m'\u001b[39m\u001b[39m_\u001b[39m\u001b[39m'\u001b[39m:\n\u001b[1;32m    451\u001b[0m     \u001b[39mraise\u001b[39;00m \u001b[39mAttributeError\u001b[39;00m(name)\n\u001b[0;32m--> 452\u001b[0m dll \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_dlltype(name)\n\u001b[1;32m    453\u001b[0m \u001b[39msetattr\u001b[39m(\u001b[39mself\u001b[39m, name, dll)\n\u001b[1;32m    454\u001b[0m \u001b[39mreturn\u001b[39;00m dll\n",
      "File \u001b[0;32m~/anaconda3/envs/bio/lib/python3.9/ctypes/__init__.py:382\u001b[0m, in \u001b[0;36mCDLL.__init__\u001b[0;34m(self, name, mode, handle, use_errno, use_last_error, winmode)\u001b[0m\n\u001b[1;32m    379\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_FuncPtr \u001b[39m=\u001b[39m _FuncPtr\n\u001b[1;32m    381\u001b[0m \u001b[39mif\u001b[39;00m handle \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[0;32m--> 382\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_handle \u001b[39m=\u001b[39m _dlopen(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_name, mode)\n\u001b[1;32m    383\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m    384\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_handle \u001b[39m=\u001b[39m handle\n",
      "\u001b[0;31mOSError\u001b[0m: /opt/intel/mkl/lib/intel64/libmkl_sequential.so: undefined symbol: mkl_sparse_z_bsr_ng_avx521_sp_mv_i4"
     ]
    }
   ],
   "source": [
    "x = tts_full_train[\"feats\"].to_numpy()\n",
    "y = tts_full_train[\"class\"].to_list()\n",
    "y = tts_full_train[\"attack\"].replace(\"A05\", \"VC\").replace(\"A06\", \"VC\").replace(\"A01\", \"TTS\").replace(\"A02\", \"TTS\").replace(\"A03\", \"TTS\").replace(\"A04\", \"TTS\").replace(\"-\",\"bonafide\")\n",
    "\n",
    "tsne = TSNE(n_components=2, verbose=1, perplexity=15, learning_rate=10)\n",
    "# tsne = TSNE(n_components=2, perplexity=15, learning_rate=10)\n",
    "z = tsne.fit_transform(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5e0ed81b-97a0-4fed-9bbf-433bee088b70",
   "metadata": {},
   "outputs": [],
   "source": [
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "e8637f2c-5dc7-4edf-96a3-08bf32d906bf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dff = pd.DataFrame()\n",
    "dff[\"y\"] = y\n",
    "dff[\"comp-1\"] = z[:,0]l\n",
    "dff[\"comp-2\"] = z[:,1]\n",
    "\n",
    "sns.scatterplot(x=\"comp-1\", y=\"comp-2\", hue=dff.y.tolist(),\n",
    "                palette=sns.color_palette(\"hls\", 2),\n",
    "                data=dff).set(title=\"\") \n",
    "plt.savefig('tts_and_bonafide_dev_2019.png',dpi=600)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0a0a5c37-9223-4a84-83b5-93cc09dfa39a",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "clf = RandomForestClassifier(n_estimators=100, random_state=0, n_jobs=-1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c9471946-f646-4ad6-97c8-f43411ea696b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight=None,\n",
       "                       criterion='gini', max_depth=None, max_features='auto',\n",
       "                       max_leaf_nodes=None, max_samples=None,\n",
       "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                       min_samples_leaf=1, min_samples_split=2,\n",
       "                       min_weight_fraction_leaf=0.0, n_estimators=100,\n",
       "                       n_jobs=-1, oob_score=False, random_state=0, verbose=0,\n",
       "                       warm_start=False)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dev_train = tts_full_train.append(tts_full_dev)\n",
    "X = dev_train['feats'].to_list()\n",
    "Y = dev_train['class'].to_numpy()\n",
    "clf.fit(X,Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "73d433ec-336e-41cc-8027-e07920f1e741",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test = tts_full_dev['feats'].to_list()\n",
    "Y_test = tts_full_dev['class'].to_numpy()\n",
    "clf.score(X_test, Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4c62a278-8dde-466f-9447-9145cd2f6cf6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 71237/71237 [04:09<00:00, 285.60it/s]\n"
     ]
    }
   ],
   "source": [
    "mypath = \"/root/biological/asvspoof2021/LA/Baseline-RawNet2-bio/feats/eval_2019/\"\n",
    "onlyfiles = [f for f in listdir(mypath) if isfile(join(mypath, f))]\n",
    "df = pd.read_csv('/root/dataset/ASVspoof/LA/ASVspoof2019_LA_cm_protocols/ASVspoof2019.LA.cm.eval.trl.txt', sep=' ')\n",
    "df.columns = [\"sid\", \"utt\", \"phy\", \"attack\", \"class\"]\n",
    "feat = pd.DataFrame()\n",
    "# feat.columns = [\"utt\", \"feats\"]\n",
    "\n",
    "for f in tqdm(onlyfiles):\n",
    "    # print(f)\n",
    "    bio_inp = torch.load(join(mypath, f)).cpu().detach().numpy()\n",
    "    feat = feat.append({\"utt\": f, \"feats\": bio_inp}, ignore_index=True)\n",
    "    # print(bio_inp)\n",
    "    \n",
    "feat.head()\n",
    "full = df.join(feat.set_index('utt'), on=\"utt\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "a433e317-ac02-4b1f-86b9-299f359f7081",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9895210110808227"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tts_full = full.loc[~full[\"attack\"].str.contains(\"A17\") & ~full[\"attack\"].str.contains(\"A18\") & ~full[\"attack\"].str.contains(\"A19\")]\n",
    "tts_full.head()\n",
    "X_eval = tts_full['feats'].to_list()\n",
    "Y_eval = tts_full['class'].to_numpy()\n",
    "clf.score(X_eval, Y_eval)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "204606cc-136c-4bde-8a07-ecb544de860d",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/ensemble/_forest.py:704: RuntimeWarning: divide by zero encountered in log\n",
      "  return np.log(proba)\n",
      "\n",
      "\n",
      "\n",
      "0it [00:00, ?it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "811it [00:00, 8109.95it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "1763it [00:00, 8485.95it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "2348it [00:00, 7474.87it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "3291it [00:00, 7968.87it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "4073it [00:00, 7921.56it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "4860it [00:00, 7905.45it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "5617it [00:00, 7800.43it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "6559it [00:00, 8223.97it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "7500it [00:00, 8546.35it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "8444it [00:01, 8794.81it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "9389it [00:01, 8980.65it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "10280it [00:01, 8625.32it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "11141it [00:01, 8248.10it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "12077it [00:01, 8551.53it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "12937it [00:01, 8252.47it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "13768it [00:01, 8102.03it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "14583it [00:01, 7378.75it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "15337it [00:02, 4756.05it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "15943it [00:02, 3719.08it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "16441it [00:02, 3256.52it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "16866it [00:02, 3145.22it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "17485it [00:02, 3689.35it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "18455it [00:02, 4531.37it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "19129it [00:02, 5024.07it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "20073it [00:03, 5844.19it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "20796it [00:03, 5942.01it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "21488it [00:03, 6151.39it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "22410it [00:03, 6832.34it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "23204it [00:03, 7128.53it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "24081it [00:03, 7551.00it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "25019it [00:03, 8018.96it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "25963it [00:03, 8396.65it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "26907it [00:03, 8682.36it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "27850it [00:03, 8892.41it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "28791it [00:04, 9041.35it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "29709it [00:04, 8639.86it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "30624it [00:04, 8786.19it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "31513it [00:04, 8297.76it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "32356it [00:04, 6290.78it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "33067it [00:04, 4352.15it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "33641it [00:05, 3478.66it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "34114it [00:05, 3076.21it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "34518it [00:05, 3011.81it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "35105it [00:05, 3526.19it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "36017it [00:05, 4321.27it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "36988it [00:05, 5184.40it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "37925it [00:05, 5986.25it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "38895it [00:05, 6762.56it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "39724it [00:06, 6940.15it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "40689it [00:06, 7577.13it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "41631it [00:06, 8049.24it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "42576it [00:06, 8423.64it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "43523it [00:06, 8710.38it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "44468it [00:06, 8919.47it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "45403it [00:06, 9041.49it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "46338it [00:06, 9130.34it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "47272it [00:06, 9191.23it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "48202it [00:07, 8666.24it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "49141it [00:07, 8870.47it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "50039it [00:07, 8438.50it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "50895it [00:07, 5974.06it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "51601it [00:07, 4049.52it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "52165it [00:08, 3279.76it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "52628it [00:08, 2638.54it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "53007it [00:08, 2813.20it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "53545it [00:08, 3282.47it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "54295it [00:08, 3948.26it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "55266it [00:08, 4803.16it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "56494it [00:08, 6392.27it/s]\u001b[A\u001b[A\u001b[A\n"
     ]
    }
   ],
   "source": [
    "# write score to file:\n",
    "sco = clf.predict_log_proba(tts_full['feats'].to_list())\n",
    "i=0\n",
    "with open(\"/root/biological/bio_ML_2019.txt\", \"a+\") as f:\n",
    "    for index, row in tqdm(tts_full.iterrows()):\n",
    "        key = row['class']\n",
    "        utt = row['utt']\n",
    "        src = row['attack']\n",
    "        # sco = clf.predict_log_proba([row['feats']])\n",
    "        f.write(\"{} {} {} {}\\n\".format(utt, src, key, sco[i][0]))\n",
    "        i=i+1\n",
    "f.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "98863611-81b0-4ad4-973a-3f5d0030bb8d",
   "metadata": {},
   "source": [
    "# SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "abe0737d-95b2-4ea5-9bf8-de91776ade5b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1.0, break_ties=False, cache_size=200, class_weight=None, coef0=0.0,\n",
       "    decision_function_shape='ovr', degree=3, gamma='scale', kernel='rbf',\n",
       "    max_iter=-1, probability=True, random_state=None, shrinking=True, tol=0.001,\n",
       "    verbose=False)"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import svm\n",
    "scl = svm.SVC(probability=True)\n",
    "scl.fit(X,Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "1fbdd4c2-33c7-4a2d-a4ea-b34e037a1501",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9861047190852126"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scl.score(X_eval, Y_eval)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "deee95dc-fde1-4953-a56d-cc6cc43af5ae",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\n",
      "0it [00:00, ?it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "927it [00:00, 9266.01it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "1375it [00:00, 7004.33it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "1704it [00:00, 4073.76it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "1996it [00:00, 3537.26it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "2484it [00:00, 3854.50it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "2858it [00:00, 3816.93it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "3580it [00:00, 4445.14it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "4552it [00:00, 5309.44it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "5527it [00:00, 6148.61it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "6494it [00:01, 6902.63it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "7467it [00:01, 7561.27it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "8317it [00:01, 7404.44it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "9124it [00:01, 7441.15it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "9915it [00:01, 7441.78it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "10692it [00:01, 7451.32it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "11475it [00:01, 7560.84it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "12400it [00:01, 7998.58it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "13219it [00:01, 7892.93it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "14022it [00:02, 7761.40it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "14808it [00:02, 4861.91it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "15436it [00:02, 4753.37it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "16099it [00:02, 5194.20it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "16864it [00:02, 5747.31it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "17532it [00:02, 5997.70it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "18259it [00:02, 6329.56it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "19000it [00:02, 6618.73it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "19742it [00:03, 6838.48it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "20471it [00:03, 6964.37it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "21361it [00:03, 7448.97it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "22282it [00:03, 7901.35it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "23096it [00:03, 7804.45it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "23893it [00:03, 7408.33it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "24650it [00:03, 7062.65it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "25593it [00:03, 7636.75it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "26486it [00:03, 7782.39it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "27282it [00:04, 4575.90it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "27909it [00:04, 4585.80it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "28487it [00:04, 4821.21it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "29402it [00:04, 5618.61it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "30341it [00:04, 6387.70it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "31094it [00:04, 6604.85it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "32033it [00:04, 7249.60it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "32834it [00:05, 7311.94it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "33619it [00:05, 7317.20it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "34525it [00:05, 7763.13it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "35335it [00:05, 7781.25it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "36137it [00:05, 7597.54it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "36915it [00:05, 7586.95it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "37846it [00:05, 8031.97it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "38679it [00:05, 8115.05it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "39502it [00:06, 5574.95it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "40175it [00:06, 4621.34it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "40745it [00:06, 4536.57it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "41542it [00:06, 5209.67it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "42487it [00:06, 6019.31it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "43409it [00:06, 6718.65it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "44183it [00:06, 6771.61it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "44932it [00:06, 6956.90it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "45865it [00:06, 7531.25it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "46789it [00:07, 7973.58it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "47630it [00:07, 8098.66it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "48549it [00:07, 8397.07it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "49477it [00:07, 8642.57it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "50410it [00:07, 8832.26it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "51322it [00:07, 8914.33it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "52224it [00:07, 6611.58it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "52979it [00:08, 4798.10it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "53596it [00:08, 4987.98it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "54287it [00:08, 5442.04it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "55015it [00:08, 5887.23it/s]\u001b[A\u001b[A\u001b[A\n",
      "\n",
      "\n",
      "56494it [00:08, 6630.67it/s]\u001b[A\u001b[A\u001b[A\n"
     ]
    }
   ],
   "source": [
    "# write score to file:\n",
    "sco = scl.predict_log_proba(tts_full['feats'].to_list())\n",
    "i=0\n",
    "with open(\"/root/biological/bio_ML_svm_2019.txt\", \"a+\") as f:\n",
    "    for index, row in tqdm(tts_full.iterrows()):\n",
    "        key = row['class']\n",
    "        utt = row['utt']\n",
    "        src = row['attack']\n",
    "        # sco = clf.predict_log_proba([row['feats']])\n",
    "        f.write(\"{} {} {} {}\\n\".format(utt, src, key, sco[i][0]))\n",
    "        i=i+1\n",
    "f.close()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "bio",
   "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": "3.9.16"
  },
  "vscode": {
   "interpreter": {
    "hash": "86c1bc152aade974d696d2553eb3be94f6dbe0748fda25ff9e7c26ed705e1ff6"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}