diff --git a/notebooks/extsream.ipynb b/notebooks/extsream_1.ipynb
similarity index 87%
rename from notebooks/extsream.ipynb
rename to notebooks/extsream_1.ipynb
index 6e8f202..7d56b25 100644
--- a/notebooks/extsream.ipynb
+++ b/notebooks/extsream_1.ipynb
@@ -9,11 +9,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [],
    "source": [
     "# Import librairies.\n",
+    "from typing import List, Tuple, Dict, Set, Union\n",
     "import glob\n",
     "import numpy as np\n",
     "import pandas as pd\n",
@@ -30,7 +31,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [
     {
@@ -45,7 +46,7 @@
        " '../data/custom_no_streaming_8/folder_1\\\\labels.csv']"
       ]
      },
-     "execution_count": 2,
+     "execution_count": 15,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -59,18 +60,21 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 55,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Concatenated DataFrame (except labels.csv):\n",
-      "(272935, 10)\n",
+      "Concatenated DataFrame (all but 'labels.csv'):\n",
+      " (272935, 10)\n",
       "\n",
       "Labels DataFrame:\n",
-      "(29, 7)\n"
+      " (29, 7)\n",
+      "\n",
+      "Total no. missing values:\n",
+      " 0\n"
      ]
     }
    ],
@@ -85,7 +89,6 @@
     "    df.insert(1, column=\"trace_id\", value=filename)\n",
     "    df.rename({\"Unnamed: 0\": \"time\"}, axis=1, inplace=True)\n",
     "    dfs.append(df)\n",
-    "\n",
     "# Concatenate all dataframes except labels.csv\n",
     "anomaly_df = pd.concat(dfs)\n",
     "\n",
@@ -93,19 +96,10 @@
     "labels_df = pd.read_csv(files[-1], index_col=0)\n",
     "\n",
     "# Print the resulting dataframes\n",
-    "print(\"Concatenated DataFrame (except labels.csv):\")\n",
-    "print(anomaly_df.shape)\n",
+    "print(\"Concatenated DataFrame (all but 'labels.csv'):\\n\", anomaly_df.shape)\n",
+    "print(\"\\nLabels DataFrame:\\n\", labels_df.shape)\n",
+    "\n",
     "\n",
-    "print(\"\\nLabels DataFrame:\")\n",
-    "print(labels_df.shape)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
-   "outputs": [],
-   "source": [
     "# Rename values in the \"anomaly_type\" column based on the specified mapping\n",
     "mapping = {\n",
     "    \"1_1\": \"bursty_input\",\n",
@@ -115,44 +109,15 @@
     "    \"3_1\": \"cpu_contention\",\n",
     "    \"3_2\": \"cpu_contention\"\n",
     "}\n",
-    "anomaly_df.insert(2, column=\"anomaly_type\", value=anomaly_df['trace_id'].replace(mapping))\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 19,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "time                                                   0\n",
-       "trace_id                                               0\n",
-       "anomaly_type                                           0\n",
-       "driver_BlockManager_memory_memUsed_MB_value            0\n",
-       "driver_jvm_heap_used_value                             0\n",
-       "avg_jvm_heap_used_value                                0\n",
-       "avg_executor_filesystem_hdfs_write_ops_value_1_diff    0\n",
-       "avg_executor_cpuTime_count_1_diff                      0\n",
-       "avg_executor_runTime_count_1_diff                      0\n",
-       "avg_executor_shuffleRecordsRead_count_1_diff           0\n",
-       "avg_executor_shuffleRecordsWritten_count_1_diff        0\n",
-       "dtype: int64"
-      ]
-     },
-     "execution_count": 19,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
+    "anomaly_df.insert(2, column=\"anomaly_type\", value=anomaly_df['trace_id'].replace(mapping))\n",
+    "\n",
     "# Missing values.\n",
-    "anomaly_df.isna().sum()"
+    "print(\"\\nTotal no. missing values:\\n\", anomaly_df.isna().sum().sum())"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 56,
    "metadata": {},
    "outputs": [
     {
@@ -178,15 +143,6 @@
        "      <th></th>\n",
        "      <th>time</th>\n",
        "      <th>trace_id</th>\n",
-       "      <th>anomaly_type</th>\n",
-       "      <th>driver_BlockManager_memory_memUsed_MB_value</th>\n",
-       "      <th>driver_jvm_heap_used_value</th>\n",
-       "      <th>avg_jvm_heap_used_value</th>\n",
-       "      <th>avg_executor_filesystem_hdfs_write_ops_value_1_diff</th>\n",
-       "      <th>avg_executor_cpuTime_count_1_diff</th>\n",
-       "      <th>avg_executor_runTime_count_1_diff</th>\n",
-       "      <th>avg_executor_shuffleRecordsRead_count_1_diff</th>\n",
-       "      <th>avg_executor_shuffleRecordsWritten_count_1_diff</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
@@ -194,268 +150,91 @@
        "      <th>0</th>\n",
        "      <td>0</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2100.0</td>\n",
-       "      <td>1.095197e+09</td>\n",
-       "      <td>2.665584e+09</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000e+00</td>\n",
-       "      <td>0.0000</td>\n",
-       "      <td>0.000</td>\n",
-       "      <td>0.000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td>1</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2134.0</td>\n",
-       "      <td>1.100890e+09</td>\n",
-       "      <td>2.690258e+09</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000e+00</td>\n",
-       "      <td>0.0000</td>\n",
-       "      <td>0.000</td>\n",
-       "      <td>0.000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td>2</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2200.0</td>\n",
-       "      <td>1.106092e+09</td>\n",
-       "      <td>2.731373e+09</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000e+00</td>\n",
-       "      <td>0.0000</td>\n",
-       "      <td>0.000</td>\n",
-       "      <td>0.000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td>3</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2265.0</td>\n",
-       "      <td>1.128138e+09</td>\n",
-       "      <td>2.116565e+09</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>4.837354e+08</td>\n",
-       "      <td>486.0000</td>\n",
-       "      <td>0.000</td>\n",
-       "      <td>8302.000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td>4</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2299.0</td>\n",
-       "      <td>1.165589e+09</td>\n",
-       "      <td>2.580707e+09</td>\n",
-       "      <td>40.000000</td>\n",
-       "      <td>1.146078e+09</td>\n",
-       "      <td>1529.0000</td>\n",
-       "      <td>25885.000</td>\n",
-       "      <td>9968.000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>...</th>\n",
        "      <td>...</td>\n",
        "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>43094</th>\n",
        "      <td>43094</td>\n",
        "      <td>3_2</td>\n",
-       "      <td>cpu_contention</td>\n",
-       "      <td>12893.0</td>\n",
-       "      <td>3.736669e+08</td>\n",
-       "      <td>1.166263e+10</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>2.516878e+09</td>\n",
-       "      <td>2639.6667</td>\n",
-       "      <td>0.000</td>\n",
-       "      <td>37973.668</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>43095</th>\n",
        "      <td>43095</td>\n",
        "      <td>3_2</td>\n",
-       "      <td>cpu_contention</td>\n",
-       "      <td>13095.0</td>\n",
-       "      <td>3.857830e+08</td>\n",
-       "      <td>9.822457e+09</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>1.470157e+09</td>\n",
-       "      <td>1509.0000</td>\n",
-       "      <td>0.000</td>\n",
-       "      <td>21917.334</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>43096</th>\n",
        "      <td>43096</td>\n",
        "      <td>3_2</td>\n",
-       "      <td>cpu_contention</td>\n",
-       "      <td>13319.0</td>\n",
-       "      <td>3.999292e+08</td>\n",
-       "      <td>9.272556e+09</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>1.807193e+09</td>\n",
-       "      <td>2253.0000</td>\n",
-       "      <td>0.000</td>\n",
-       "      <td>25897.666</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>43097</th>\n",
        "      <td>43097</td>\n",
        "      <td>3_2</td>\n",
-       "      <td>cpu_contention</td>\n",
-       "      <td>13598.0</td>\n",
-       "      <td>4.153047e+08</td>\n",
-       "      <td>8.529498e+09</td>\n",
-       "      <td>6.000000</td>\n",
-       "      <td>1.821398e+09</td>\n",
-       "      <td>1899.6666</td>\n",
-       "      <td>28229.666</td>\n",
-       "      <td>28906.666</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>43098</th>\n",
        "      <td>43098</td>\n",
        "      <td>3_2</td>\n",
-       "      <td>cpu_contention</td>\n",
-       "      <td>13802.0</td>\n",
-       "      <td>4.516347e+08</td>\n",
-       "      <td>8.042026e+09</td>\n",
-       "      <td>20.666666</td>\n",
-       "      <td>5.880769e+08</td>\n",
-       "      <td>833.6667</td>\n",
-       "      <td>114175.000</td>\n",
-       "      <td>0.000</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
-       "<p>272935 rows × 11 columns</p>\n",
+       "<p>272935 rows × 2 columns</p>\n",
        "</div>"
       ],
       "text/plain": [
-       "        time trace_id    anomaly_type  \\\n",
-       "0          0      1_1    bursty_input   \n",
-       "1          1      1_1    bursty_input   \n",
-       "2          2      1_1    bursty_input   \n",
-       "3          3      1_1    bursty_input   \n",
-       "4          4      1_1    bursty_input   \n",
-       "...      ...      ...             ...   \n",
-       "43094  43094      3_2  cpu_contention   \n",
-       "43095  43095      3_2  cpu_contention   \n",
-       "43096  43096      3_2  cpu_contention   \n",
-       "43097  43097      3_2  cpu_contention   \n",
-       "43098  43098      3_2  cpu_contention   \n",
-       "\n",
-       "       driver_BlockManager_memory_memUsed_MB_value  \\\n",
-       "0                                           2100.0   \n",
-       "1                                           2134.0   \n",
-       "2                                           2200.0   \n",
-       "3                                           2265.0   \n",
-       "4                                           2299.0   \n",
-       "...                                            ...   \n",
-       "43094                                      12893.0   \n",
-       "43095                                      13095.0   \n",
-       "43096                                      13319.0   \n",
-       "43097                                      13598.0   \n",
-       "43098                                      13802.0   \n",
+       "        time trace_id\n",
+       "0          0      1_1\n",
+       "1          1      1_1\n",
+       "2          2      1_1\n",
+       "3          3      1_1\n",
+       "4          4      1_1\n",
+       "...      ...      ...\n",
+       "43094  43094      3_2\n",
+       "43095  43095      3_2\n",
+       "43096  43096      3_2\n",
+       "43097  43097      3_2\n",
+       "43098  43098      3_2\n",
        "\n",
-       "       driver_jvm_heap_used_value  avg_jvm_heap_used_value  \\\n",
-       "0                    1.095197e+09             2.665584e+09   \n",
-       "1                    1.100890e+09             2.690258e+09   \n",
-       "2                    1.106092e+09             2.731373e+09   \n",
-       "3                    1.128138e+09             2.116565e+09   \n",
-       "4                    1.165589e+09             2.580707e+09   \n",
-       "...                           ...                      ...   \n",
-       "43094                3.736669e+08             1.166263e+10   \n",
-       "43095                3.857830e+08             9.822457e+09   \n",
-       "43096                3.999292e+08             9.272556e+09   \n",
-       "43097                4.153047e+08             8.529498e+09   \n",
-       "43098                4.516347e+08             8.042026e+09   \n",
-       "\n",
-       "       avg_executor_filesystem_hdfs_write_ops_value_1_diff  \\\n",
-       "0                                               0.000000     \n",
-       "1                                               0.000000     \n",
-       "2                                               0.000000     \n",
-       "3                                               0.000000     \n",
-       "4                                              40.000000     \n",
-       "...                                                  ...     \n",
-       "43094                                           0.000000     \n",
-       "43095                                           0.000000     \n",
-       "43096                                           0.000000     \n",
-       "43097                                           6.000000     \n",
-       "43098                                          20.666666     \n",
-       "\n",
-       "       avg_executor_cpuTime_count_1_diff  avg_executor_runTime_count_1_diff  \\\n",
-       "0                           0.000000e+00                             0.0000   \n",
-       "1                           0.000000e+00                             0.0000   \n",
-       "2                           0.000000e+00                             0.0000   \n",
-       "3                           4.837354e+08                           486.0000   \n",
-       "4                           1.146078e+09                          1529.0000   \n",
-       "...                                  ...                                ...   \n",
-       "43094                       2.516878e+09                          2639.6667   \n",
-       "43095                       1.470157e+09                          1509.0000   \n",
-       "43096                       1.807193e+09                          2253.0000   \n",
-       "43097                       1.821398e+09                          1899.6666   \n",
-       "43098                       5.880769e+08                           833.6667   \n",
-       "\n",
-       "       avg_executor_shuffleRecordsRead_count_1_diff  \\\n",
-       "0                                             0.000   \n",
-       "1                                             0.000   \n",
-       "2                                             0.000   \n",
-       "3                                             0.000   \n",
-       "4                                         25885.000   \n",
-       "...                                             ...   \n",
-       "43094                                         0.000   \n",
-       "43095                                         0.000   \n",
-       "43096                                         0.000   \n",
-       "43097                                     28229.666   \n",
-       "43098                                    114175.000   \n",
-       "\n",
-       "       avg_executor_shuffleRecordsWritten_count_1_diff  \n",
-       "0                                                0.000  \n",
-       "1                                                0.000  \n",
-       "2                                                0.000  \n",
-       "3                                             8302.000  \n",
-       "4                                             9968.000  \n",
-       "...                                                ...  \n",
-       "43094                                        37973.668  \n",
-       "43095                                        21917.334  \n",
-       "43096                                        25897.666  \n",
-       "43097                                        28906.666  \n",
-       "43098                                            0.000  \n",
-       "\n",
-       "[272935 rows x 11 columns]"
+       "[272935 rows x 2 columns]"
       ]
      },
-     "execution_count": 10,
+     "execution_count": 56,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "anomaly_df"
+    "anomaly_df[[\"time\", \"trace_id\"]]."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 45,
    "metadata": {},
    "outputs": [
     {
@@ -479,364 +258,390 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
+       "      <th>time</th>\n",
        "      <th>trace_id</th>\n",
+       "      <th>anomaly_type</th>\n",
+       "      <th>driver_BlockManager_memory_memUsed_MB_value</th>\n",
+       "      <th>driver_jvm_heap_used_value</th>\n",
+       "      <th>avg_jvm_heap_used_value</th>\n",
+       "      <th>avg_executor_filesystem_hdfs_write_ops_value_1_diff</th>\n",
+       "      <th>avg_executor_cpuTime_count_1_diff</th>\n",
+       "      <th>avg_executor_runTime_count_1_diff</th>\n",
+       "      <th>avg_executor_shuffleRecordsRead_count_1_diff</th>\n",
+       "      <th>avg_executor_shuffleRecordsWritten_count_1_diff</th>\n",
        "      <th>ano_id</th>\n",
        "      <th>ref_start</th>\n",
        "      <th>ref_end</th>\n",
        "      <th>ano_start</th>\n",
        "      <th>ano_end</th>\n",
        "      <th>ano_type</th>\n",
+       "      <th>_merge</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>0</th>\n",
+       "      <th>25482</th>\n",
+       "      <td>4247</td>\n",
        "      <td>1_1</td>\n",
+       "      <td>bursty_input</td>\n",
+       "      <td>2189.0</td>\n",
+       "      <td>367144260.0</td>\n",
+       "      <td>3.028880e+09</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000e+00</td>\n",
+       "      <td>0.0000</td>\n",
+       "      <td>0.000</td>\n",
+       "      <td>0.0000</td>\n",
        "      <td>0</td>\n",
        "      <td>4247</td>\n",
        "      <td>6193</td>\n",
        "      <td>6193</td>\n",
        "      <td>7166</td>\n",
        "      <td>bursty_input</td>\n",
+       "      <td>both</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>1</th>\n",
+       "      <th>25488</th>\n",
+       "      <td>4248</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>11425</td>\n",
-       "      <td>13393</td>\n",
-       "      <td>13393</td>\n",
-       "      <td>14377</td>\n",
        "      <td>bursty_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>1_1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>18659</td>\n",
-       "      <td>20593</td>\n",
-       "      <td>20593</td>\n",
-       "      <td>21560</td>\n",
+       "      <td>2255.0</td>\n",
+       "      <td>635267260.0</td>\n",
+       "      <td>4.005025e+09</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>4.942382e+08</td>\n",
+       "      <td>495.0000</td>\n",
+       "      <td>0.000</td>\n",
+       "      <td>14532.5000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>4247</td>\n",
+       "      <td>6193</td>\n",
+       "      <td>6193</td>\n",
+       "      <td>7166</td>\n",
        "      <td>bursty_input</td>\n",
+       "      <td>both</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>3</th>\n",
+       "      <th>25494</th>\n",
+       "      <td>4249</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>3</td>\n",
-       "      <td>25849</td>\n",
-       "      <td>27793</td>\n",
-       "      <td>27793</td>\n",
-       "      <td>28765</td>\n",
        "      <td>bursty_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>1_1</td>\n",
-       "      <td>4</td>\n",
-       "      <td>33049</td>\n",
-       "      <td>34993</td>\n",
-       "      <td>34993</td>\n",
-       "      <td>35965</td>\n",
+       "      <td>2288.0</td>\n",
+       "      <td>735523600.0</td>\n",
+       "      <td>4.224833e+09</td>\n",
+       "      <td>34.000000</td>\n",
+       "      <td>8.029332e+08</td>\n",
+       "      <td>1147.0000</td>\n",
+       "      <td>34207.500</td>\n",
+       "      <td>10806.5000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>4247</td>\n",
+       "      <td>6193</td>\n",
+       "      <td>6193</td>\n",
+       "      <td>7166</td>\n",
        "      <td>bursty_input</td>\n",
+       "      <td>both</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>5</th>\n",
+       "      <th>25500</th>\n",
+       "      <td>4250</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>5</td>\n",
-       "      <td>40257</td>\n",
-       "      <td>42193</td>\n",
-       "      <td>42193</td>\n",
-       "      <td>43161</td>\n",
        "      <td>bursty_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>6</th>\n",
-       "      <td>1_2</td>\n",
+       "      <td>2142.0</td>\n",
+       "      <td>760737500.0</td>\n",
+       "      <td>2.231513e+09</td>\n",
+       "      <td>6.000000</td>\n",
+       "      <td>6.237813e+07</td>\n",
+       "      <td>490.0000</td>\n",
+       "      <td>7250.500</td>\n",
+       "      <td>0.0000</td>\n",
        "      <td>0</td>\n",
-       "      <td>5055</td>\n",
-       "      <td>6989</td>\n",
-       "      <td>6989</td>\n",
-       "      <td>7956</td>\n",
-       "      <td>bursty_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>7</th>\n",
-       "      <td>1_2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>13025</td>\n",
-       "      <td>14989</td>\n",
-       "      <td>14989</td>\n",
-       "      <td>15971</td>\n",
-       "      <td>bursty_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>8</th>\n",
-       "      <td>1_2</td>\n",
-       "      <td>2</td>\n",
-       "      <td>21043</td>\n",
-       "      <td>22989</td>\n",
-       "      <td>22989</td>\n",
-       "      <td>23962</td>\n",
-       "      <td>bursty_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>9</th>\n",
-       "      <td>1_2</td>\n",
-       "      <td>3</td>\n",
-       "      <td>29033</td>\n",
-       "      <td>30989</td>\n",
-       "      <td>30989</td>\n",
-       "      <td>31967</td>\n",
+       "      <td>4247</td>\n",
+       "      <td>6193</td>\n",
+       "      <td>6193</td>\n",
+       "      <td>7166</td>\n",
        "      <td>bursty_input</td>\n",
+       "      <td>both</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>10</th>\n",
-       "      <td>1_2</td>\n",
-       "      <td>4</td>\n",
-       "      <td>37045</td>\n",
-       "      <td>38989</td>\n",
-       "      <td>38989</td>\n",
-       "      <td>39961</td>\n",
+       "      <th>25506</th>\n",
+       "      <td>4251</td>\n",
+       "      <td>1_1</td>\n",
        "      <td>bursty_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>11</th>\n",
-       "      <td>2_1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>6954</td>\n",
-       "      <td>8908</td>\n",
-       "      <td>8908</td>\n",
-       "      <td>9885</td>\n",
-       "      <td>stalled_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>12</th>\n",
-       "      <td>2_1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>16856</td>\n",
-       "      <td>18808</td>\n",
-       "      <td>18808</td>\n",
-       "      <td>19784</td>\n",
-       "      <td>stalled_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>13</th>\n",
-       "      <td>2_1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>26756</td>\n",
-       "      <td>28708</td>\n",
-       "      <td>28708</td>\n",
-       "      <td>29684</td>\n",
-       "      <td>stalled_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>14</th>\n",
-       "      <td>2_1</td>\n",
-       "      <td>3</td>\n",
-       "      <td>36659</td>\n",
-       "      <td>38609</td>\n",
-       "      <td>38609</td>\n",
-       "      <td>39584</td>\n",
-       "      <td>stalled_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>15</th>\n",
-       "      <td>2_2</td>\n",
+       "      <td>2125.0</td>\n",
+       "      <td>770470140.0</td>\n",
+       "      <td>2.237279e+09</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000e+00</td>\n",
+       "      <td>0.0000</td>\n",
+       "      <td>0.000</td>\n",
+       "      <td>0.0000</td>\n",
        "      <td>0</td>\n",
-       "      <td>8771</td>\n",
-       "      <td>10699</td>\n",
-       "      <td>10699</td>\n",
-       "      <td>11663</td>\n",
-       "      <td>stalled_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>16</th>\n",
-       "      <td>2_2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>20451</td>\n",
-       "      <td>22399</td>\n",
-       "      <td>22399</td>\n",
-       "      <td>23373</td>\n",
-       "      <td>stalled_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>17</th>\n",
-       "      <td>2_2</td>\n",
-       "      <td>2</td>\n",
-       "      <td>32154</td>\n",
-       "      <td>34100</td>\n",
-       "      <td>34100</td>\n",
-       "      <td>35073</td>\n",
-       "      <td>stalled_input</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>18</th>\n",
-       "      <td>2_2</td>\n",
-       "      <td>3</td>\n",
-       "      <td>44024</td>\n",
-       "      <td>45800</td>\n",
-       "      <td>45800</td>\n",
-       "      <td>46688</td>\n",
-       "      <td>stalled_input</td>\n",
+       "      <td>4247</td>\n",
+       "      <td>6193</td>\n",
+       "      <td>6193</td>\n",
+       "      <td>7166</td>\n",
+       "      <td>bursty_input</td>\n",
+       "      <td>both</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>19</th>\n",
-       "      <td>3_1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3035</td>\n",
-       "      <td>3965</td>\n",
-       "      <td>3965</td>\n",
-       "      <td>4430</td>\n",
-       "      <td>cpu_contention</td>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>20</th>\n",
-       "      <td>3_1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>5205</td>\n",
-       "      <td>7125</td>\n",
-       "      <td>7125</td>\n",
-       "      <td>8085</td>\n",
+       "      <th>1286118</th>\n",
+       "      <td>36724</td>\n",
+       "      <td>3_2</td>\n",
        "      <td>cpu_contention</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>21</th>\n",
-       "      <td>3_1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>11442</td>\n",
-       "      <td>12548</td>\n",
-       "      <td>12548</td>\n",
-       "      <td>13101</td>\n",
+       "      <td>12790.0</td>\n",
+       "      <td>522976540.0</td>\n",
+       "      <td>9.267732e+09</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>1.557150e+09</td>\n",
+       "      <td>1589.0000</td>\n",
+       "      <td>0.000</td>\n",
+       "      <td>29848.6660</td>\n",
+       "      <td>4</td>\n",
+       "      <td>31802</td>\n",
+       "      <td>35086</td>\n",
+       "      <td>35086</td>\n",
+       "      <td>36728</td>\n",
        "      <td>cpu_contention</td>\n",
+       "      <td>both</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>22</th>\n",
-       "      <td>3_1</td>\n",
-       "      <td>3</td>\n",
-       "      <td>17332</td>\n",
-       "      <td>19980</td>\n",
-       "      <td>19980</td>\n",
-       "      <td>21304</td>\n",
+       "      <th>1286123</th>\n",
+       "      <td>36725</td>\n",
+       "      <td>3_2</td>\n",
        "      <td>cpu_contention</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>23</th>\n",
-       "      <td>3_1</td>\n",
+       "      <td>13045.0</td>\n",
+       "      <td>534622100.0</td>\n",
+       "      <td>8.986438e+09</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>1.908520e+09</td>\n",
+       "      <td>2291.0000</td>\n",
+       "      <td>0.000</td>\n",
+       "      <td>30756.3340</td>\n",
        "      <td>4</td>\n",
-       "      <td>31947</td>\n",
-       "      <td>35035</td>\n",
-       "      <td>35035</td>\n",
-       "      <td>36579</td>\n",
+       "      <td>31802</td>\n",
+       "      <td>35086</td>\n",
+       "      <td>35086</td>\n",
+       "      <td>36728</td>\n",
        "      <td>cpu_contention</td>\n",
+       "      <td>both</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>24</th>\n",
+       "      <th>1286128</th>\n",
+       "      <td>36726</td>\n",
        "      <td>3_2</td>\n",
-       "      <td>0</td>\n",
-       "      <td>3050</td>\n",
-       "      <td>4016</td>\n",
-       "      <td>4016</td>\n",
-       "      <td>4499</td>\n",
        "      <td>cpu_contention</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>25</th>\n",
-       "      <td>3_2</td>\n",
-       "      <td>1</td>\n",
-       "      <td>5288</td>\n",
-       "      <td>7176</td>\n",
-       "      <td>7176</td>\n",
-       "      <td>8120</td>\n",
+       "      <td>13209.0</td>\n",
+       "      <td>549762800.0</td>\n",
+       "      <td>8.434711e+09</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>2.480250e+09</td>\n",
+       "      <td>2544.3333</td>\n",
+       "      <td>0.000</td>\n",
+       "      <td>38853.0000</td>\n",
+       "      <td>4</td>\n",
+       "      <td>31802</td>\n",
+       "      <td>35086</td>\n",
+       "      <td>35086</td>\n",
+       "      <td>36728</td>\n",
        "      <td>cpu_contention</td>\n",
+       "      <td>both</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>26</th>\n",
+       "      <th>1286133</th>\n",
+       "      <td>36727</td>\n",
        "      <td>3_2</td>\n",
-       "      <td>2</td>\n",
-       "      <td>17455</td>\n",
-       "      <td>20031</td>\n",
-       "      <td>20031</td>\n",
-       "      <td>21319</td>\n",
        "      <td>cpu_contention</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>27</th>\n",
-       "      <td>3_2</td>\n",
-       "      <td>3</td>\n",
-       "      <td>26522</td>\n",
-       "      <td>28060</td>\n",
-       "      <td>28060</td>\n",
-       "      <td>28829</td>\n",
+       "      <td>13486.0</td>\n",
+       "      <td>574724700.0</td>\n",
+       "      <td>9.416742e+09</td>\n",
+       "      <td>22.666666</td>\n",
+       "      <td>6.127938e+08</td>\n",
+       "      <td>721.0000</td>\n",
+       "      <td>117025.000</td>\n",
+       "      <td>1511.3334</td>\n",
+       "      <td>4</td>\n",
+       "      <td>31802</td>\n",
+       "      <td>35086</td>\n",
+       "      <td>35086</td>\n",
+       "      <td>36728</td>\n",
        "      <td>cpu_contention</td>\n",
+       "      <td>both</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>28</th>\n",
+       "      <th>1286138</th>\n",
+       "      <td>36728</td>\n",
        "      <td>3_2</td>\n",
+       "      <td>cpu_contention</td>\n",
+       "      <td>13784.0</td>\n",
+       "      <td>606572860.0</td>\n",
+       "      <td>8.501003e+09</td>\n",
+       "      <td>4.000000</td>\n",
+       "      <td>1.040760e+08</td>\n",
+       "      <td>134.0000</td>\n",
+       "      <td>20870.666</td>\n",
+       "      <td>0.0000</td>\n",
        "      <td>4</td>\n",
        "      <td>31802</td>\n",
        "      <td>35086</td>\n",
        "      <td>35086</td>\n",
        "      <td>36728</td>\n",
        "      <td>cpu_contention</td>\n",
+       "      <td>both</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
+       "<p>85177 rows × 18 columns</p>\n",
        "</div>"
       ],
       "text/plain": [
-       "   trace_id  ano_id  ref_start  ref_end  ano_start  ano_end        ano_type\n",
-       "0       1_1       0       4247     6193       6193     7166    bursty_input\n",
-       "1       1_1       1      11425    13393      13393    14377    bursty_input\n",
-       "2       1_1       2      18659    20593      20593    21560    bursty_input\n",
-       "3       1_1       3      25849    27793      27793    28765    bursty_input\n",
-       "4       1_1       4      33049    34993      34993    35965    bursty_input\n",
-       "5       1_1       5      40257    42193      42193    43161    bursty_input\n",
-       "6       1_2       0       5055     6989       6989     7956    bursty_input\n",
-       "7       1_2       1      13025    14989      14989    15971    bursty_input\n",
-       "8       1_2       2      21043    22989      22989    23962    bursty_input\n",
-       "9       1_2       3      29033    30989      30989    31967    bursty_input\n",
-       "10      1_2       4      37045    38989      38989    39961    bursty_input\n",
-       "11      2_1       0       6954     8908       8908     9885   stalled_input\n",
-       "12      2_1       1      16856    18808      18808    19784   stalled_input\n",
-       "13      2_1       2      26756    28708      28708    29684   stalled_input\n",
-       "14      2_1       3      36659    38609      38609    39584   stalled_input\n",
-       "15      2_2       0       8771    10699      10699    11663   stalled_input\n",
-       "16      2_2       1      20451    22399      22399    23373   stalled_input\n",
-       "17      2_2       2      32154    34100      34100    35073   stalled_input\n",
-       "18      2_2       3      44024    45800      45800    46688   stalled_input\n",
-       "19      3_1       0       3035     3965       3965     4430  cpu_contention\n",
-       "20      3_1       1       5205     7125       7125     8085  cpu_contention\n",
-       "21      3_1       2      11442    12548      12548    13101  cpu_contention\n",
-       "22      3_1       3      17332    19980      19980    21304  cpu_contention\n",
-       "23      3_1       4      31947    35035      35035    36579  cpu_contention\n",
-       "24      3_2       0       3050     4016       4016     4499  cpu_contention\n",
-       "25      3_2       1       5288     7176       7176     8120  cpu_contention\n",
-       "26      3_2       2      17455    20031      20031    21319  cpu_contention\n",
-       "27      3_2       3      26522    28060      28060    28829  cpu_contention\n",
-       "28      3_2       4      31802    35086      35086    36728  cpu_contention"
+       "          time trace_id    anomaly_type  \\\n",
+       "25482     4247      1_1    bursty_input   \n",
+       "25488     4248      1_1    bursty_input   \n",
+       "25494     4249      1_1    bursty_input   \n",
+       "25500     4250      1_1    bursty_input   \n",
+       "25506     4251      1_1    bursty_input   \n",
+       "...        ...      ...             ...   \n",
+       "1286118  36724      3_2  cpu_contention   \n",
+       "1286123  36725      3_2  cpu_contention   \n",
+       "1286128  36726      3_2  cpu_contention   \n",
+       "1286133  36727      3_2  cpu_contention   \n",
+       "1286138  36728      3_2  cpu_contention   \n",
+       "\n",
+       "         driver_BlockManager_memory_memUsed_MB_value  \\\n",
+       "25482                                         2189.0   \n",
+       "25488                                         2255.0   \n",
+       "25494                                         2288.0   \n",
+       "25500                                         2142.0   \n",
+       "25506                                         2125.0   \n",
+       "...                                              ...   \n",
+       "1286118                                      12790.0   \n",
+       "1286123                                      13045.0   \n",
+       "1286128                                      13209.0   \n",
+       "1286133                                      13486.0   \n",
+       "1286138                                      13784.0   \n",
+       "\n",
+       "         driver_jvm_heap_used_value  avg_jvm_heap_used_value  \\\n",
+       "25482                   367144260.0             3.028880e+09   \n",
+       "25488                   635267260.0             4.005025e+09   \n",
+       "25494                   735523600.0             4.224833e+09   \n",
+       "25500                   760737500.0             2.231513e+09   \n",
+       "25506                   770470140.0             2.237279e+09   \n",
+       "...                             ...                      ...   \n",
+       "1286118                 522976540.0             9.267732e+09   \n",
+       "1286123                 534622100.0             8.986438e+09   \n",
+       "1286128                 549762800.0             8.434711e+09   \n",
+       "1286133                 574724700.0             9.416742e+09   \n",
+       "1286138                 606572860.0             8.501003e+09   \n",
+       "\n",
+       "         avg_executor_filesystem_hdfs_write_ops_value_1_diff  \\\n",
+       "25482                                             0.000000     \n",
+       "25488                                             0.000000     \n",
+       "25494                                            34.000000     \n",
+       "25500                                             6.000000     \n",
+       "25506                                             0.000000     \n",
+       "...                                                    ...     \n",
+       "1286118                                           0.000000     \n",
+       "1286123                                           0.000000     \n",
+       "1286128                                           0.000000     \n",
+       "1286133                                          22.666666     \n",
+       "1286138                                           4.000000     \n",
+       "\n",
+       "         avg_executor_cpuTime_count_1_diff  avg_executor_runTime_count_1_diff  \\\n",
+       "25482                         0.000000e+00                             0.0000   \n",
+       "25488                         4.942382e+08                           495.0000   \n",
+       "25494                         8.029332e+08                          1147.0000   \n",
+       "25500                         6.237813e+07                           490.0000   \n",
+       "25506                         0.000000e+00                             0.0000   \n",
+       "...                                    ...                                ...   \n",
+       "1286118                       1.557150e+09                          1589.0000   \n",
+       "1286123                       1.908520e+09                          2291.0000   \n",
+       "1286128                       2.480250e+09                          2544.3333   \n",
+       "1286133                       6.127938e+08                           721.0000   \n",
+       "1286138                       1.040760e+08                           134.0000   \n",
+       "\n",
+       "         avg_executor_shuffleRecordsRead_count_1_diff  \\\n",
+       "25482                                           0.000   \n",
+       "25488                                           0.000   \n",
+       "25494                                       34207.500   \n",
+       "25500                                        7250.500   \n",
+       "25506                                           0.000   \n",
+       "...                                               ...   \n",
+       "1286118                                         0.000   \n",
+       "1286123                                         0.000   \n",
+       "1286128                                         0.000   \n",
+       "1286133                                    117025.000   \n",
+       "1286138                                     20870.666   \n",
+       "\n",
+       "         avg_executor_shuffleRecordsWritten_count_1_diff  ano_id  ref_start  \\\n",
+       "25482                                             0.0000       0       4247   \n",
+       "25488                                         14532.5000       0       4247   \n",
+       "25494                                         10806.5000       0       4247   \n",
+       "25500                                             0.0000       0       4247   \n",
+       "25506                                             0.0000       0       4247   \n",
+       "...                                                  ...     ...        ...   \n",
+       "1286118                                       29848.6660       4      31802   \n",
+       "1286123                                       30756.3340       4      31802   \n",
+       "1286128                                       38853.0000       4      31802   \n",
+       "1286133                                        1511.3334       4      31802   \n",
+       "1286138                                           0.0000       4      31802   \n",
+       "\n",
+       "         ref_end  ano_start  ano_end        ano_type _merge  \n",
+       "25482       6193       6193     7166    bursty_input   both  \n",
+       "25488       6193       6193     7166    bursty_input   both  \n",
+       "25494       6193       6193     7166    bursty_input   both  \n",
+       "25500       6193       6193     7166    bursty_input   both  \n",
+       "25506       6193       6193     7166    bursty_input   both  \n",
+       "...          ...        ...      ...             ...    ...  \n",
+       "1286118    35086      35086    36728  cpu_contention   both  \n",
+       "1286123    35086      35086    36728  cpu_contention   both  \n",
+       "1286128    35086      35086    36728  cpu_contention   both  \n",
+       "1286133    35086      35086    36728  cpu_contention   both  \n",
+       "1286138    35086      35086    36728  cpu_contention   both  \n",
+       "\n",
+       "[85177 rows x 18 columns]"
       ]
      },
-     "execution_count": 11,
+     "execution_count": 45,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "labels_df"
+    "pd.merge(\n",
+    "    anomaly_df,\n",
+    "    labels_df,\n",
+    "    how='inner',\n",
+    "    on='trace_id',\n",
+    "    suffixes=('_anomaly', '_labels'),\n",
+    "    indicator=True\n",
+    ").loc[lambda x: x['time'].between(x['ref_start'], x['ref_end']) | x['time'].between(x['ano_start'], x['ano_end'])]\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "bursty_input1_df = anomaly_df[anomaly_df[\"trace_id\"] == \"1_1\"]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 59,
    "metadata": {},
    "outputs": [
     {
@@ -860,286 +665,104 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
-       "      <th>time</th>\n",
        "      <th>trace_id</th>\n",
-       "      <th>anomaly_type</th>\n",
-       "      <th>driver_BlockManager_memory_memUsed_MB_value</th>\n",
-       "      <th>driver_jvm_heap_used_value</th>\n",
-       "      <th>avg_jvm_heap_used_value</th>\n",
-       "      <th>avg_executor_filesystem_hdfs_write_ops_value_1_diff</th>\n",
-       "      <th>avg_executor_cpuTime_count_1_diff</th>\n",
-       "      <th>avg_executor_runTime_count_1_diff</th>\n",
-       "      <th>avg_executor_shuffleRecordsRead_count_1_diff</th>\n",
-       "      <th>avg_executor_shuffleRecordsWritten_count_1_diff</th>\n",
+       "      <th>ano_id</th>\n",
+       "      <th>ref_start</th>\n",
+       "      <th>ref_end</th>\n",
+       "      <th>ano_start</th>\n",
+       "      <th>ano_end</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
-       "      <td>0</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2100.0</td>\n",
-       "      <td>1.095197e+09</td>\n",
-       "      <td>2.665584e+09</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.000000e+00</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>4247</td>\n",
+       "      <td>6193</td>\n",
+       "      <td>6193</td>\n",
+       "      <td>7166</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
-       "      <td>1</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2134.0</td>\n",
-       "      <td>1.100890e+09</td>\n",
-       "      <td>2.690258e+09</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.000000e+00</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>11425</td>\n",
+       "      <td>13393</td>\n",
+       "      <td>13393</td>\n",
+       "      <td>14377</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
-       "      <td>2</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2200.0</td>\n",
-       "      <td>1.106092e+09</td>\n",
-       "      <td>2.731373e+09</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.000000e+00</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
+       "      <td>2</td>\n",
+       "      <td>18659</td>\n",
+       "      <td>20593</td>\n",
+       "      <td>20593</td>\n",
+       "      <td>21560</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
-       "      <td>3</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2265.0</td>\n",
-       "      <td>1.128138e+09</td>\n",
-       "      <td>2.116565e+09</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>4.837354e+08</td>\n",
-       "      <td>486.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>8302.0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>25849</td>\n",
+       "      <td>27793</td>\n",
+       "      <td>27793</td>\n",
+       "      <td>28765</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
-       "      <td>4</td>\n",
        "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2299.0</td>\n",
-       "      <td>1.165589e+09</td>\n",
-       "      <td>2.580707e+09</td>\n",
-       "      <td>40.0</td>\n",
-       "      <td>1.146078e+09</td>\n",
-       "      <td>1529.0</td>\n",
-       "      <td>25885.0</td>\n",
-       "      <td>9968.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>46686</th>\n",
-       "      <td>46686</td>\n",
-       "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2095.0</td>\n",
-       "      <td>4.809436e+08</td>\n",
-       "      <td>9.870312e+09</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.000000e+00</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>46687</th>\n",
-       "      <td>46687</td>\n",
-       "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2143.0</td>\n",
-       "      <td>4.904456e+08</td>\n",
-       "      <td>9.910990e+09</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.000000e+00</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>46688</th>\n",
-       "      <td>46688</td>\n",
-       "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2208.0</td>\n",
-       "      <td>5.183927e+08</td>\n",
-       "      <td>8.766686e+09</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>4.356684e+08</td>\n",
-       "      <td>451.5</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>7001.5</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>46689</th>\n",
-       "      <td>46689</td>\n",
-       "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2258.0</td>\n",
-       "      <td>5.530916e+08</td>\n",
-       "      <td>8.766686e+09</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.000000e+00</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>46690</th>\n",
-       "      <td>46690</td>\n",
-       "      <td>1_1</td>\n",
-       "      <td>bursty_input</td>\n",
-       "      <td>2258.0</td>\n",
-       "      <td>5.830545e+08</td>\n",
-       "      <td>8.766686e+09</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.000000e+00</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.0</td>\n",
+       "      <td>4</td>\n",
+       "      <td>33049</td>\n",
+       "      <td>34993</td>\n",
+       "      <td>34993</td>\n",
+       "      <td>35965</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
-       "<p>46691 rows × 11 columns</p>\n",
        "</div>"
       ],
       "text/plain": [
-       "        time trace_id  anomaly_type  \\\n",
-       "0          0      1_1  bursty_input   \n",
-       "1          1      1_1  bursty_input   \n",
-       "2          2      1_1  bursty_input   \n",
-       "3          3      1_1  bursty_input   \n",
-       "4          4      1_1  bursty_input   \n",
-       "...      ...      ...           ...   \n",
-       "46686  46686      1_1  bursty_input   \n",
-       "46687  46687      1_1  bursty_input   \n",
-       "46688  46688      1_1  bursty_input   \n",
-       "46689  46689      1_1  bursty_input   \n",
-       "46690  46690      1_1  bursty_input   \n",
-       "\n",
-       "       driver_BlockManager_memory_memUsed_MB_value  \\\n",
-       "0                                           2100.0   \n",
-       "1                                           2134.0   \n",
-       "2                                           2200.0   \n",
-       "3                                           2265.0   \n",
-       "4                                           2299.0   \n",
-       "...                                            ...   \n",
-       "46686                                       2095.0   \n",
-       "46687                                       2143.0   \n",
-       "46688                                       2208.0   \n",
-       "46689                                       2258.0   \n",
-       "46690                                       2258.0   \n",
-       "\n",
-       "       driver_jvm_heap_used_value  avg_jvm_heap_used_value  \\\n",
-       "0                    1.095197e+09             2.665584e+09   \n",
-       "1                    1.100890e+09             2.690258e+09   \n",
-       "2                    1.106092e+09             2.731373e+09   \n",
-       "3                    1.128138e+09             2.116565e+09   \n",
-       "4                    1.165589e+09             2.580707e+09   \n",
-       "...                           ...                      ...   \n",
-       "46686                4.809436e+08             9.870312e+09   \n",
-       "46687                4.904456e+08             9.910990e+09   \n",
-       "46688                5.183927e+08             8.766686e+09   \n",
-       "46689                5.530916e+08             8.766686e+09   \n",
-       "46690                5.830545e+08             8.766686e+09   \n",
-       "\n",
-       "       avg_executor_filesystem_hdfs_write_ops_value_1_diff  \\\n",
-       "0                                                    0.0     \n",
-       "1                                                    0.0     \n",
-       "2                                                    0.0     \n",
-       "3                                                    0.0     \n",
-       "4                                                   40.0     \n",
-       "...                                                  ...     \n",
-       "46686                                                0.0     \n",
-       "46687                                                0.0     \n",
-       "46688                                                0.0     \n",
-       "46689                                                0.0     \n",
-       "46690                                                0.0     \n",
-       "\n",
-       "       avg_executor_cpuTime_count_1_diff  avg_executor_runTime_count_1_diff  \\\n",
-       "0                           0.000000e+00                                0.0   \n",
-       "1                           0.000000e+00                                0.0   \n",
-       "2                           0.000000e+00                                0.0   \n",
-       "3                           4.837354e+08                              486.0   \n",
-       "4                           1.146078e+09                             1529.0   \n",
-       "...                                  ...                                ...   \n",
-       "46686                       0.000000e+00                                0.0   \n",
-       "46687                       0.000000e+00                                0.0   \n",
-       "46688                       4.356684e+08                              451.5   \n",
-       "46689                       0.000000e+00                                0.0   \n",
-       "46690                       0.000000e+00                                0.0   \n",
-       "\n",
-       "       avg_executor_shuffleRecordsRead_count_1_diff  \\\n",
-       "0                                               0.0   \n",
-       "1                                               0.0   \n",
-       "2                                               0.0   \n",
-       "3                                               0.0   \n",
-       "4                                           25885.0   \n",
-       "...                                             ...   \n",
-       "46686                                           0.0   \n",
-       "46687                                           0.0   \n",
-       "46688                                           0.0   \n",
-       "46689                                           0.0   \n",
-       "46690                                           0.0   \n",
-       "\n",
-       "       avg_executor_shuffleRecordsWritten_count_1_diff  \n",
-       "0                                                  0.0  \n",
-       "1                                                  0.0  \n",
-       "2                                                  0.0  \n",
-       "3                                               8302.0  \n",
-       "4                                               9968.0  \n",
-       "...                                                ...  \n",
-       "46686                                              0.0  \n",
-       "46687                                              0.0  \n",
-       "46688                                           7001.5  \n",
-       "46689                                              0.0  \n",
-       "46690                                              0.0  \n",
-       "\n",
-       "[46691 rows x 11 columns]"
+       "  trace_id  ano_id  ref_start  ref_end  ano_start  ano_end\n",
+       "0      1_1       0       4247     6193       6193     7166\n",
+       "1      1_1       1      11425    13393      13393    14377\n",
+       "2      1_1       2      18659    20593      20593    21560\n",
+       "3      1_1       3      25849    27793      27793    28765\n",
+       "4      1_1       4      33049    34993      34993    35965"
       ]
      },
-     "execution_count": 21,
+     "execution_count": 59,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "bursty_input1_df"
+    "labels_df.iloc[:, :-1].head()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 40,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "['1_1']\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Take one example\n",
+    "bursty_input1_df = anomaly_df[anomaly_df[\"trace_id\"] == \"1_1\"]\n",
+    "print(bursty_input1_df[\"trace_id\"].unique())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
    "metadata": {},
    "outputs": [
     {
@@ -1201,11 +824,37 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 24,
    "metadata": {},
    "outputs": [],
    "source": [
-    "#---CODE HERE----"
+    "# #---CODE HERE----\n",
+    "# def sufficient_features_space(\n",
+    "#         data: pd.DataFrame | np.ndarray,\n",
+    "#         time_feature: str | int,\n",
+    "#         ref_start: int, \n",
+    "#         ref_end: int, \n",
+    "#         ano_start: int, \n",
+    "#         ano_end: int) -> List[str] | List[int]:\n",
+    "    \n",
+    "#     if isinstance(data, pd.DataFrame):\n",
+    "#         mask_ref = data[time_feature].between(ref_start, ref_end)\n",
+    "#         mask_ano = data[time_feature].between(ano_start, ano_end)\n",
+    "#         mask_intervals = mask_ref and mask_ano\n",
+    "#         relevant_data = data[mask_intervals]\n",
+    "#         return [col for col in relevant_data.columns if relevant_data[col].var() > 10e-16]\n",
+    "#     else:\n",
+    "#         raise AttributeError(\"`data` is not a dataframe\")\n",
+    "#     # pass"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# sufficient_features_space(bursty_input1_df, time_feature=)"
    ]
   },
   {
@@ -1250,27 +899,117 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [],
    "source": [
-    "#---CODE HERE----"
+    "#---CODE HERE----\n",
+    "def filter_feature(data: pd.DataFrame, time: str, feature: str, ref_start: int, ref_end: int, ano_start: int, ano_end: int):\n",
+    "    mask_ref = data[time].between(ref_start, ref_end)\n",
+    "    mask_ano = data[time].between(ano_start, ano_end)\n",
+    "    TSA = data.loc[mask_ano, [time, feature]].set_index(time)\n",
+    "    TSR = data.loc[mask_ref, [time, feature]].set_index(time)\n",
+    "    return TSA, TSR"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "4247 6193 6193 7166\n",
+      "(1947, 1) time\n",
+      "4248    494238180.0\n",
+      "6193    423371900.0\n",
+      "Name: avg_executor_cpuTime_count_1_diff, dtype: float64\n",
+      "(974, 1) time\n",
+      "6194    1.444361e+09\n",
+      "7166    0.000000e+00\n",
+      "Name: avg_executor_cpuTime_count_1_diff, dtype: float64\n"
+     ]
+    }
+   ],
+   "source": [
+    "ref_start, ref_end, ano_start, ano_end = labels_df.loc[\n",
+    "    0, [\"ref_start\", \"ref_end\", \"ano_start\", \"ano_end\"]\n",
+    "] #TODO: Replace \"0\" by a selected \"ano_id\" given a \"trace_id\"\n",
+    "\n",
+    "TSA, TSR = filter_feature(data=bursty_input1_df,\n",
+    "               time=\"time\",\n",
+    "               feature=\"avg_executor_cpuTime_count_1_diff\",\n",
+    "               ref_start=ref_start,\n",
+    "               ref_end=ref_end,\n",
+    "               ano_start=ano_start,\n",
+    "               ano_end=ano_end)\n",
+    "\n",
+    "print(ref_start, ref_end, ano_start, ano_end)\n",
+    "print(TSR.shape, TSR.iloc[[1,-1], 0])\n",
+    "print(TSA.shape, TSA.iloc[[1,-1], 0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def class_entropy(TSA, TSR):\n",
+    "    \"\"\"\n",
+    "    Calculate the class entropy between two time series.\n",
+    "\n",
+    "    Parameters:\n",
+    "    - TSA: Time series belonging to the abnormal class.\n",
+    "    - TSR: Time series belonging to the reference class.\n",
+    "\n",
+    "    Returns:\n",
+    "    - HClass: Class entropy.\n",
+    "    \"\"\"\n",
+    "    # Calculate probabilities\n",
+    "    pA = len(TSA) / (len(TSA) + len(TSR))\n",
+    "    pR = len(TSR) / (len(TSA) + len(TSR))\n",
+    "\n",
+    "    return - pA * np.log(pA) - pR * np.log(pR)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.6365932382975314\n"
+     ]
+    }
+   ],
+   "source": [
+    "class_ent = class_entropy(TSA, TSR)\n",
+    "print(class_ent)"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Segmentation entropy (???)\n",
+    "### Segmentation entropy\n",
     "\n",
     "**Input**\n",
-    "* Time series $TS$ of a feature $f$: $(X_t)_{t=1,\\dots,T}$. \n",
+    "* Time series $TS$ of a feature $f$: $(X_t)_{t=1,\\dots,T}$.\n",
     "* Anomaly time annotations: $A_S = (a_s)_{s=1,\\dots, S}  \\subset \\{1, \\dots, T\\}$  \n",
     "\n",
     "Example: In the sequence, AANNNNANNN. The anomaly time anotations would be (1,2,7). We are given only the different time intervals from which we have to derive the different anomaly time anotation and then the segment (this is just an heuristic, not necessarily the way it would be implemented).\n",
     "\n",
+    "\n",
     "**Compute**\n",
-    "* The segments. A segment is a (contiguous) sequence of point with the same label (anomaly or not anomaly)\n",
+    "* Sort the feature time series values in increasing order.\n",
+    "* Tag each feature value depending of if it appears during the anormal interval only (YELLOW), the reference interval only (RED), or in both interval (BLUE).\n",
+    "* The segments. A segment is a (contiguous) sequence of point with the same label (anomaly or not anomaly).\n",
     "* The segmentation entropy of the feature $f$:\n",
     "$$\n",
     "    H_{Segmentation}(f) = \\sum_{i=1}^{n} p_i \\cdot \\log\\left(\\frac{1}{p_i}\\right)\n",
@@ -1283,7 +1022,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "#---CODE HERE----"
+    "#---CODE HERE----\n",
+    "def segmentation_entropy():\n",
+    "    pass"
    ]
   },
   {
@@ -1307,7 +1048,16 @@
    "execution_count": null,
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "def reorder_mixed_segment():\n",
+    "    pass\n",
+    "\n",
+    "def segment_penalty(): # ???\n",
+    "    pass\n",
+    "\n",
+    "def penalized_segmentation_entropy():\n",
+    "    pass"
+   ]
   },
   {
    "cell_type": "markdown",
@@ -1316,12 +1066,31 @@
     "### Normalization by feature size"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "    D(f) = \\dfrac{H_{Class}(f)}{H^+_{Segmentation}(f)}\n",
+    "$$"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "def single_reward_function(feature):\n",
+    "    return class_entropy(feature) / penalized_segmentation_entropy(feature)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Testing `single_reward_function`"
+   ]
   },
   {
    "cell_type": "code",