data-exploration.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "5815ea87",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:48:08.872893Z",
     "iopub.status.busy": "2024-02-02T22:48:08.872166Z",
     "iopub.status.idle": "2024-02-02T22:48:08.876402Z",
     "shell.execute_reply": "2024-02-02T22:48:08.875747Z",
     "shell.execute_reply.started": "2024-02-02T22:48:08.872861Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import functools\n",
    "import json\n",
    "\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from videoannotator import config as cfg, data, io"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "74582bac-6464-4542-b5e2-9e503dfb76fa",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:20.054898Z",
     "iopub.status.busy": "2024-02-02T22:36:20.054453Z",
     "iopub.status.idle": "2024-02-02T22:36:20.269944Z",
     "shell.execute_reply": "2024-02-02T22:36:20.269377Z",
     "shell.execute_reply.started": "2024-02-02T22:36:20.054875Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# label to label group mapping\n",
    "l2g = {l: g for g, ls in cfg.LABEL_GROUPS.items() for l in ls}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "9f605f8a-85cb-4c1a-bb07-0f5c4575e1de",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:51:23.126507Z",
     "iopub.status.busy": "2024-02-02T22:51:23.125750Z",
     "iopub.status.idle": "2024-02-02T22:51:23.866686Z",
     "shell.execute_reply": "2024-02-02T22:51:23.865916Z",
     "shell.execute_reply.started": "2024-02-02T22:51:23.126475Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5.0M\t/root/data/videoannotator/agg\n",
      "23M\t/root/data/videoannotator/ave/validation\n",
      "134M\t/root/data/videoannotator/ave\n",
      "17M\t/root/data/videoannotator/checkpoints\n",
      "11M\t/root/data/videoannotator/cmp\n",
      "1.2G\t/root/data/videoannotator/\n"
     ]
    }
   ],
   "source": [
    "!du -h /root/data/videoannotator/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "a8248802-f737-42fc-ab5f-35bf27397ea1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:50:41.366175Z",
     "iopub.status.busy": "2024-02-02T22:50:41.365413Z",
     "iopub.status.idle": "2024-02-02T22:50:41.523155Z",
     "shell.execute_reply": "2024-02-02T22:50:41.522352Z",
     "shell.execute_reply.started": "2024-02-02T22:50:41.366144Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total 1.1G\n",
      "drwxr-xr-x 2 root root    0 May 31  2023 agg\n",
      "drwxr-xr-x 3 root root    0 May 11  2023 ave\n",
      "drwxr-xr-x 2 root root    0 May 10  2023 checkpoints\n",
      "drwxr-xr-x 2 root root    0 Jun  7  2023 cmp\n",
      "-rw-r--r-- 1 root root 985M May 12  2023 embeddings.h5\n",
      "-rw-r--r-- 1 root root  11K May 31  2023 queries.json\n",
      "-rw-r--r-- 1 root root  66M May 10  2023 shot-data.csv\n",
      "-rw-r--r-- 1 root root 647K May 19  2023 text-embeddings.json\n"
     ]
    }
   ],
   "source": [
    "!ls -lh /root/data/videoannotator/"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "f801647c-4f26-45eb-9117-2cafe2e24468",
   "metadata": {},
   "source": [
    "## list of all shots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "8c9f9b22-1d2e-46c6-9276-e169fbbc58ee",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:44:00.547222Z",
     "iopub.status.busy": "2024-02-02T22:44:00.546654Z",
     "iopub.status.idle": "2024-02-02T22:44:01.578945Z",
     "shell.execute_reply": "2024-02-02T22:44:01.578318Z",
     "shell.execute_reply.started": "2024-02-02T22:44:00.547192Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "504030"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_shots = pd.read_csv(io.PATHS_STATIC.shot_data)\n",
    "len(df_shots)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "54e3a348-97dd-4931-b8b0-37c33015c8cb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:21.430535Z",
     "iopub.status.busy": "2024-02-02T22:36:21.430057Z",
     "iopub.status.idle": "2024-02-02T22:36:21.443111Z",
     "shell.execute_reply": "2024-02-02T22:36:21.442556Z",
     "shell.execute_reply.started": "2024-02-02T22:36:21.430511Z"
    },
    "tags": []
   },
   "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>key</th>\n",
       "      <th>yt_id</th>\n",
       "      <th>desc</th>\n",
       "      <th>genre</th>\n",
       "      <th>frame_in</th>\n",
       "      <th>frame_out</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>fps</th>\n",
       "      <th>cluster_index</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>HnnBvemHrWA|0</td>\n",
       "      <td>HnnBvemHrWA</td>\n",
       "      <td>Frankie and Johnny - He Just Asked Her Out</td>\n",
       "      <td>Comedy</td>\n",
       "      <td>0</td>\n",
       "      <td>82</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.420311</td>\n",
       "      <td>23.974427</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>gOJJm_cSRds|3987</td>\n",
       "      <td>gOJJm_cSRds</td>\n",
       "      <td>The Wild Bunch - Bank Shootout</td>\n",
       "      <td>Action</td>\n",
       "      <td>3987</td>\n",
       "      <td>4007</td>\n",
       "      <td>166.302203</td>\n",
       "      <td>167.136425</td>\n",
       "      <td>23.974427</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mDViU8OSRkA|2222</td>\n",
       "      <td>mDViU8OSRkA</td>\n",
       "      <td>Non-Stop - How's Your Daughter? Scene</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2222</td>\n",
       "      <td>2284</td>\n",
       "      <td>92.675917</td>\n",
       "      <td>95.261833</td>\n",
       "      <td>23.976024</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>egB-SG97EcI|805</td>\n",
       "      <td>egB-SG97EcI</td>\n",
       "      <td>Kalifornia - I Think I Gotta Kill You Scene</td>\n",
       "      <td>Crime</td>\n",
       "      <td>805</td>\n",
       "      <td>825</td>\n",
       "      <td>33.575190</td>\n",
       "      <td>34.409357</td>\n",
       "      <td>23.976037</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>uGsWYV2bWAc|1013</td>\n",
       "      <td>uGsWYV2bWAc</td>\n",
       "      <td>Kill Bill: Vol. 1 - The Bride vs. Gogo</td>\n",
       "      <td>Action</td>\n",
       "      <td>1013</td>\n",
       "      <td>1041</td>\n",
       "      <td>42.250542</td>\n",
       "      <td>43.418375</td>\n",
       "      <td>23.976024</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                key        yt_id                                         desc  \\\n",
       "0     HnnBvemHrWA|0  HnnBvemHrWA   Frankie and Johnny - He Just Asked Her Out   \n",
       "1  gOJJm_cSRds|3987  gOJJm_cSRds               The Wild Bunch - Bank Shootout   \n",
       "2  mDViU8OSRkA|2222  mDViU8OSRkA        Non-Stop - How's Your Daughter? Scene   \n",
       "3   egB-SG97EcI|805  egB-SG97EcI  Kalifornia - I Think I Gotta Kill You Scene   \n",
       "4  uGsWYV2bWAc|1013  uGsWYV2bWAc       Kill Bill: Vol. 1 - The Bride vs. Gogo   \n",
       "\n",
       "    genre  frame_in  frame_out       start         end        fps  \\\n",
       "0  Comedy         0         82    0.000000    3.420311  23.974427   \n",
       "1  Action      3987       4007  166.302203  167.136425  23.974427   \n",
       "2     NaN      2222       2284   92.675917   95.261833  23.976024   \n",
       "3   Crime       805        825   33.575190   34.409357  23.976037   \n",
       "4  Action      1013       1041   42.250542   43.418375  23.976024   \n",
       "\n",
       "   cluster_index  \n",
       "0              8  \n",
       "1              6  \n",
       "2              3  \n",
       "3              0  \n",
       "4              7  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_shots.head()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "e71be14f-291e-4fd7-89c0-a7ec970953dd",
   "metadata": {},
   "source": [
    "## getting individual annotations for a label\n",
    "this shows the aggregated annotation per shot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "8d6ab560-ac0b-473d-8ac7-11ad5a3bf7ad",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:38:41.273204Z",
     "iopub.status.busy": "2024-02-02T22:38:41.272474Z",
     "iopub.status.idle": "2024-02-02T22:38:41.278791Z",
     "shell.execute_reply": "2024-02-02T22:38:41.278216Z",
     "shell.execute_reply.started": "2024-02-02T22:38:41.273174Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LabeledDataset(label=closeup, cnt=460, pos_cnt=223, neg_cnt=237, pos_rate=0.48)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label = 'closeup'\n",
    "lds = data.get_aggregate_labeled_dataset(label=label)\n",
    "lds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "978d5a92-f35d-4545-b5a9-92b6235d70e5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:40:41.417579Z",
     "iopub.status.busy": "2024-02-02T22:40:41.416873Z",
     "iopub.status.idle": "2024-02-02T22:40:41.421750Z",
     "shell.execute_reply": "2024-02-02T22:40:41.421170Z",
     "shell.execute_reply.started": "2024-02-02T22:40:41.417542Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(223, 237)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# lds.keys_pos contains positive annotations and lds.keys_neg contains the negative ones\n",
    "# note that \"key\" specifies a unique shot (see the `key` column in `df_shots`)\n",
    "# for instance closeup has 223 positive and 237 negative instances\n",
    "len(lds.keys_pos), len(lds.keys_neg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "8e5807e9-f4ef-465c-8168-fe1ed1efb198",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:41:10.238955Z",
     "iopub.status.busy": "2024-02-02T22:41:10.238353Z",
     "iopub.status.idle": "2024-02-02T22:41:10.243138Z",
     "shell.execute_reply": "2024-02-02T22:41:10.242567Z",
     "shell.execute_reply.started": "2024-02-02T22:41:10.238924Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['09MjH5hbF5Y|1369', '09MjH5hbF5Y|1722', '0QNpXi4k5eU|3005']"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# here's the first 3 positive keys\n",
    "lds.keys_pos[:3]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "a9eb1055-23b2-4684-8213-6cff56572489",
   "metadata": {},
   "source": [
    "### want the granular view?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "108aaa3d-81fc-4772-a41b-77ca5453d625",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:49:55.569420Z",
     "iopub.status.busy": "2024-02-02T22:49:55.568976Z",
     "iopub.status.idle": "2024-02-02T22:49:55.605174Z",
     "shell.execute_reply": "2024-02-02T22:49:55.604571Z",
     "shell.execute_reply.started": "2024-02-02T22:49:55.569389Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'B'"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label = 'closeup'\n",
    "path = f'{io.PATHS_STATIC._path_base}/agg/{label}.json'\n",
    "details = json.load(open(path))\n",
    "details['primary']  # annotator B is the primary for closeup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "eab78d2f-f72f-432c-991e-d86089ca5304",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:49:58.403478Z",
     "iopub.status.busy": "2024-02-02T22:49:58.402752Z",
     "iopub.status.idle": "2024-02-02T22:49:58.407565Z",
     "shell.execute_reply": "2024-02-02T22:49:58.406969Z",
     "shell.execute_reply.started": "2024-02-02T22:49:58.403448Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['A', 'C']"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# A and C are secondaries\n",
    "details['secondaries']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "18e11756-46e9-4dab-9bcd-99807a681596",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:49:14.911241Z",
     "iopub.status.busy": "2024-02-02T22:49:14.910492Z",
     "iopub.status.idle": "2024-02-02T22:49:14.915467Z",
     "shell.execute_reply": "2024-02-02T22:49:14.914831Z",
     "shell.execute_reply.started": "2024-02-02T22:49:14.911210Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "460"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(details['data'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "bea2e930-78e5-4909-9114-902618209205",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:49:21.830211Z",
     "iopub.status.busy": "2024-02-02T22:49:21.829465Z",
     "iopub.status.idle": "2024-02-02T22:49:21.835215Z",
     "shell.execute_reply": "2024-02-02T22:49:21.834630Z",
     "shell.execute_reply.started": "2024-02-02T22:49:21.830179Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'key': 'ddrQKAMzUGI|2015',\n",
       "  'primary': True,\n",
       "  'secondaries': {'A': True, 'C': True},\n",
       "  'agg': True,\n",
       "  'pos_cnt': 3},\n",
       " {'key': 'XhwgtlCns34|867',\n",
       "  'primary': True,\n",
       "  'secondaries': {'A': True, 'C': True},\n",
       "  'agg': True,\n",
       "  'pos_cnt': 3},\n",
       " {'key': 'LxmnOxCk3dM|4393',\n",
       "  'primary': True,\n",
       "  'secondaries': {'A': True, 'C': True},\n",
       "  'agg': True,\n",
       "  'pos_cnt': 3}]"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "details['data'][:3]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "6df80369-9e77-42eb-b02c-7203c24178d0",
   "metadata": {},
   "source": [
    "## labeled data summary stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2b2417ca-1978-47fc-9467-04ab2f5bdb2a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:21.444341Z",
     "iopub.status.busy": "2024-02-02T22:36:21.443896Z",
     "iopub.status.idle": "2024-02-02T22:36:21.607795Z",
     "shell.execute_reply": "2024-02-02T22:36:21.607214Z",
     "shell.execute_reply.started": "2024-02-02T22:36:21.444318Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def stats(l):\n",
    "    lds = data.get_aggregate_labeled_dataset(l)\n",
    "    cmp = data.CompDataset(l)\n",
    "    rand = cmp.get_ordered_dataset_random().lds\n",
    "    zs = cmp.get_ordered_dataset_zero_shot().lds\n",
    "    return dict(\n",
    "        label=l,\n",
    "        cnt=len(lds),\n",
    "        cnt_pos=lds.pos_cnt,\n",
    "        pos_rate=lds.pos_rate,\n",
    "        keys=set(lds.keys),\n",
    "        cnt_zs=len(zs),\n",
    "        cnt_zs_pos=zs.pos_cnt,\n",
    "        zs_pos_rate=zs.pos_rate,\n",
    "        cnt_rand=len(rand),\n",
    "        cnt_rand_pos=rand.pos_cnt,\n",
    "        agreement=data.get_labeled_dataset_agreement(label=l),\n",
    "        keys_unique=set(lds.keys) | set(zs.keys) | set(rand.keys),\n",
    "        unique_cnt=len(set(lds.keys) | set(zs.keys) | set(rand.keys)),\n",
    "        group=l2g[l],\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e6e4b957-b62c-4a2c-a68b-e0c70782446e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:21.611607Z",
     "iopub.status.busy": "2024-02-02T22:36:21.611162Z",
     "iopub.status.idle": "2024-02-02T22:36:22.260592Z",
     "shell.execute_reply": "2024-02-02T22:36:22.259986Z",
     "shell.execute_reply.started": "2024-02-02T22:36:21.611583Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING 2024-02-02 22:36:21,851 data.py(64) LabeledDataset(label=timelapse, cnt=1000, pos_cnt=0, neg_cnt=1000, pos_rate=0.00) has no positive annotations.\n",
      "WARNING 2024-02-02 22:36:21,926 data.py(64) LabeledDataset(label=alcohol, cnt=1000, pos_cnt=0, neg_cnt=1000, pos_rate=0.00) has no positive annotations.\n",
      "WARNING 2024-02-02 22:36:22,121 data.py(64) LabeledDataset(label=shutter-shot, cnt=1000, pos_cnt=0, neg_cnt=1000, pos_rate=0.00) has no positive annotations.\n",
      "WARNING 2024-02-02 22:36:22,122 data.py(64) LabeledDataset(label=shutter-shot, cnt=1000, pos_cnt=0, neg_cnt=1000, pos_rate=0.00) has no positive annotations.\n",
      "WARNING 2024-02-02 22:36:22,180 data.py(64) LabeledDataset(label=jump-scare, cnt=1000, pos_cnt=0, neg_cnt=1000, pos_rate=0.00) has no positive annotations.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>label</th>\n",
       "      <th>cnt</th>\n",
       "      <th>cnt_pos</th>\n",
       "      <th>pos_rate</th>\n",
       "      <th>cnt_zs</th>\n",
       "      <th>cnt_zs_pos</th>\n",
       "      <th>zs_pos_rate</th>\n",
       "      <th>cnt_rand</th>\n",
       "      <th>cnt_rand_pos</th>\n",
       "      <th>agreement</th>\n",
       "      <th>unique_cnt</th>\n",
       "      <th>group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>happy</td>\n",
       "      <td>1000</td>\n",
       "      <td>287</td>\n",
       "      <td>0.287000</td>\n",
       "      <td>1000</td>\n",
       "      <td>689</td>\n",
       "      <td>0.689</td>\n",
       "      <td>1000</td>\n",
       "      <td>33</td>\n",
       "      <td>0.843000</td>\n",
       "      <td>2927</td>\n",
       "      <td>emotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>zoom</td>\n",
       "      <td>1000</td>\n",
       "      <td>66</td>\n",
       "      <td>0.066000</td>\n",
       "      <td>1000</td>\n",
       "      <td>66</td>\n",
       "      <td>0.066</td>\n",
       "      <td>1000</td>\n",
       "      <td>26</td>\n",
       "      <td>0.890000</td>\n",
       "      <td>2929</td>\n",
       "      <td>motion</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>static-shot</td>\n",
       "      <td>1000</td>\n",
       "      <td>451</td>\n",
       "      <td>0.451000</td>\n",
       "      <td>1000</td>\n",
       "      <td>16</td>\n",
       "      <td>0.016</td>\n",
       "      <td>1000</td>\n",
       "      <td>73</td>\n",
       "      <td>0.881000</td>\n",
       "      <td>2995</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>low-angle</td>\n",
       "      <td>272</td>\n",
       "      <td>102</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>1000</td>\n",
       "      <td>166</td>\n",
       "      <td>0.166</td>\n",
       "      <td>1000</td>\n",
       "      <td>35</td>\n",
       "      <td>0.852941</td>\n",
       "      <td>2253</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>drama</td>\n",
       "      <td>599</td>\n",
       "      <td>203</td>\n",
       "      <td>0.338898</td>\n",
       "      <td>1000</td>\n",
       "      <td>590</td>\n",
       "      <td>0.590</td>\n",
       "      <td>1000</td>\n",
       "      <td>129</td>\n",
       "      <td>0.646077</td>\n",
       "      <td>2563</td>\n",
       "      <td>genres</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>violence</td>\n",
       "      <td>575</td>\n",
       "      <td>193</td>\n",
       "      <td>0.335652</td>\n",
       "      <td>1000</td>\n",
       "      <td>807</td>\n",
       "      <td>0.807</td>\n",
       "      <td>1000</td>\n",
       "      <td>109</td>\n",
       "      <td>0.787826</td>\n",
       "      <td>2546</td>\n",
       "      <td>sensitivities</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>cowboy-shot</td>\n",
       "      <td>1000</td>\n",
       "      <td>373</td>\n",
       "      <td>0.373000</td>\n",
       "      <td>1000</td>\n",
       "      <td>121</td>\n",
       "      <td>0.121</td>\n",
       "      <td>1000</td>\n",
       "      <td>17</td>\n",
       "      <td>0.848000</td>\n",
       "      <td>2934</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>sci-fi</td>\n",
       "      <td>768</td>\n",
       "      <td>257</td>\n",
       "      <td>0.334635</td>\n",
       "      <td>1000</td>\n",
       "      <td>777</td>\n",
       "      <td>0.777</td>\n",
       "      <td>1000</td>\n",
       "      <td>21</td>\n",
       "      <td>0.842448</td>\n",
       "      <td>2712</td>\n",
       "      <td>genres</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>day</td>\n",
       "      <td>249</td>\n",
       "      <td>124</td>\n",
       "      <td>0.497992</td>\n",
       "      <td>1000</td>\n",
       "      <td>505</td>\n",
       "      <td>0.505</td>\n",
       "      <td>1000</td>\n",
       "      <td>370</td>\n",
       "      <td>0.975904</td>\n",
       "      <td>2233</td>\n",
       "      <td>time_location</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>intimacy</td>\n",
       "      <td>1000</td>\n",
       "      <td>336</td>\n",
       "      <td>0.336000</td>\n",
       "      <td>1000</td>\n",
       "      <td>473</td>\n",
       "      <td>0.473</td>\n",
       "      <td>1000</td>\n",
       "      <td>15</td>\n",
       "      <td>0.904000</td>\n",
       "      <td>2909</td>\n",
       "      <td>sensitivities</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>car-chase</td>\n",
       "      <td>1000</td>\n",
       "      <td>403</td>\n",
       "      <td>0.403000</td>\n",
       "      <td>1000</td>\n",
       "      <td>408</td>\n",
       "      <td>0.408</td>\n",
       "      <td>1000</td>\n",
       "      <td>8</td>\n",
       "      <td>0.778000</td>\n",
       "      <td>2921</td>\n",
       "      <td>events_actions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>establishing-shots</td>\n",
       "      <td>938</td>\n",
       "      <td>449</td>\n",
       "      <td>0.478678</td>\n",
       "      <td>1000</td>\n",
       "      <td>870</td>\n",
       "      <td>0.870</td>\n",
       "      <td>1000</td>\n",
       "      <td>57</td>\n",
       "      <td>0.772921</td>\n",
       "      <td>2863</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>timelapse</td>\n",
       "      <td>1003</td>\n",
       "      <td>6</td>\n",
       "      <td>0.005982</td>\n",
       "      <td>1000</td>\n",
       "      <td>12</td>\n",
       "      <td>0.012</td>\n",
       "      <td>1000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.987039</td>\n",
       "      <td>2944</td>\n",
       "      <td>motion</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>high-angle</td>\n",
       "      <td>484</td>\n",
       "      <td>225</td>\n",
       "      <td>0.464876</td>\n",
       "      <td>1000</td>\n",
       "      <td>598</td>\n",
       "      <td>0.598</td>\n",
       "      <td>1000</td>\n",
       "      <td>38</td>\n",
       "      <td>0.845041</td>\n",
       "      <td>2430</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>smoking</td>\n",
       "      <td>1000</td>\n",
       "      <td>217</td>\n",
       "      <td>0.217000</td>\n",
       "      <td>1000</td>\n",
       "      <td>194</td>\n",
       "      <td>0.194</td>\n",
       "      <td>1000</td>\n",
       "      <td>4</td>\n",
       "      <td>0.953000</td>\n",
       "      <td>2844</td>\n",
       "      <td>sensitivities</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>overhead-shot</td>\n",
       "      <td>513</td>\n",
       "      <td>132</td>\n",
       "      <td>0.257310</td>\n",
       "      <td>1000</td>\n",
       "      <td>342</td>\n",
       "      <td>0.342</td>\n",
       "      <td>1000</td>\n",
       "      <td>8</td>\n",
       "      <td>0.740741</td>\n",
       "      <td>2481</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>laughter</td>\n",
       "      <td>1000</td>\n",
       "      <td>307</td>\n",
       "      <td>0.307000</td>\n",
       "      <td>1000</td>\n",
       "      <td>227</td>\n",
       "      <td>0.227</td>\n",
       "      <td>1000</td>\n",
       "      <td>5</td>\n",
       "      <td>0.795000</td>\n",
       "      <td>2888</td>\n",
       "      <td>emotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>eye-level</td>\n",
       "      <td>1000</td>\n",
       "      <td>675</td>\n",
       "      <td>0.675000</td>\n",
       "      <td>1000</td>\n",
       "      <td>482</td>\n",
       "      <td>0.482</td>\n",
       "      <td>1000</td>\n",
       "      <td>555</td>\n",
       "      <td>0.642000</td>\n",
       "      <td>2986</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>drugs</td>\n",
       "      <td>1000</td>\n",
       "      <td>107</td>\n",
       "      <td>0.107000</td>\n",
       "      <td>1000</td>\n",
       "      <td>119</td>\n",
       "      <td>0.119</td>\n",
       "      <td>1000</td>\n",
       "      <td>5</td>\n",
       "      <td>0.871000</td>\n",
       "      <td>2860</td>\n",
       "      <td>sensitivities</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>fight</td>\n",
       "      <td>1000</td>\n",
       "      <td>285</td>\n",
       "      <td>0.285000</td>\n",
       "      <td>1000</td>\n",
       "      <td>715</td>\n",
       "      <td>0.715</td>\n",
       "      <td>1000</td>\n",
       "      <td>17</td>\n",
       "      <td>0.868000</td>\n",
       "      <td>2938</td>\n",
       "      <td>events_actions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>alcohol</td>\n",
       "      <td>1000</td>\n",
       "      <td>268</td>\n",
       "      <td>0.268000</td>\n",
       "      <td>1000</td>\n",
       "      <td>269</td>\n",
       "      <td>0.269</td>\n",
       "      <td>1000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.780000</td>\n",
       "      <td>2859</td>\n",
       "      <td>sensitivities</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>anger</td>\n",
       "      <td>1000</td>\n",
       "      <td>215</td>\n",
       "      <td>0.215000</td>\n",
       "      <td>1000</td>\n",
       "      <td>257</td>\n",
       "      <td>0.257</td>\n",
       "      <td>1000</td>\n",
       "      <td>21</td>\n",
       "      <td>0.624000</td>\n",
       "      <td>2876</td>\n",
       "      <td>emotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>two-shot</td>\n",
       "      <td>402</td>\n",
       "      <td>246</td>\n",
       "      <td>0.611940</td>\n",
       "      <td>1000</td>\n",
       "      <td>37</td>\n",
       "      <td>0.037</td>\n",
       "      <td>1000</td>\n",
       "      <td>76</td>\n",
       "      <td>0.875622</td>\n",
       "      <td>2399</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>pan</td>\n",
       "      <td>1000</td>\n",
       "      <td>154</td>\n",
       "      <td>0.154000</td>\n",
       "      <td>1000</td>\n",
       "      <td>28</td>\n",
       "      <td>0.028</td>\n",
       "      <td>1000</td>\n",
       "      <td>12</td>\n",
       "      <td>0.870000</td>\n",
       "      <td>2976</td>\n",
       "      <td>motion</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>object</td>\n",
       "      <td>351</td>\n",
       "      <td>136</td>\n",
       "      <td>0.387464</td>\n",
       "      <td>1000</td>\n",
       "      <td>729</td>\n",
       "      <td>0.729</td>\n",
       "      <td>1000</td>\n",
       "      <td>68</td>\n",
       "      <td>0.823362</td>\n",
       "      <td>2347</td>\n",
       "      <td>focus</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>animal</td>\n",
       "      <td>420</td>\n",
       "      <td>166</td>\n",
       "      <td>0.395238</td>\n",
       "      <td>1000</td>\n",
       "      <td>769</td>\n",
       "      <td>0.769</td>\n",
       "      <td>1000</td>\n",
       "      <td>10</td>\n",
       "      <td>0.940476</td>\n",
       "      <td>2384</td>\n",
       "      <td>focus</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>slow-motion</td>\n",
       "      <td>1000</td>\n",
       "      <td>64</td>\n",
       "      <td>0.064000</td>\n",
       "      <td>1000</td>\n",
       "      <td>117</td>\n",
       "      <td>0.117</td>\n",
       "      <td>1000</td>\n",
       "      <td>7</td>\n",
       "      <td>0.939000</td>\n",
       "      <td>2915</td>\n",
       "      <td>motion</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>over-the-shoulder-shot</td>\n",
       "      <td>637</td>\n",
       "      <td>295</td>\n",
       "      <td>0.463108</td>\n",
       "      <td>1000</td>\n",
       "      <td>16</td>\n",
       "      <td>0.016</td>\n",
       "      <td>1000</td>\n",
       "      <td>38</td>\n",
       "      <td>0.897959</td>\n",
       "      <td>2633</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>handheld</td>\n",
       "      <td>1000</td>\n",
       "      <td>340</td>\n",
       "      <td>0.340000</td>\n",
       "      <td>1000</td>\n",
       "      <td>91</td>\n",
       "      <td>0.091</td>\n",
       "      <td>1000</td>\n",
       "      <td>52</td>\n",
       "      <td>0.813000</td>\n",
       "      <td>2992</td>\n",
       "      <td>motion</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>point-of-view-shot</td>\n",
       "      <td>1000</td>\n",
       "      <td>201</td>\n",
       "      <td>0.201000</td>\n",
       "      <td>1000</td>\n",
       "      <td>50</td>\n",
       "      <td>0.050</td>\n",
       "      <td>1000</td>\n",
       "      <td>1</td>\n",
       "      <td>0.844000</td>\n",
       "      <td>2949</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>single-shot</td>\n",
       "      <td>357</td>\n",
       "      <td>159</td>\n",
       "      <td>0.445378</td>\n",
       "      <td>1000</td>\n",
       "      <td>117</td>\n",
       "      <td>0.117</td>\n",
       "      <td>1000</td>\n",
       "      <td>271</td>\n",
       "      <td>0.795518</td>\n",
       "      <td>2350</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>insert-shot</td>\n",
       "      <td>387</td>\n",
       "      <td>194</td>\n",
       "      <td>0.501292</td>\n",
       "      <td>1000</td>\n",
       "      <td>178</td>\n",
       "      <td>0.178</td>\n",
       "      <td>1000</td>\n",
       "      <td>60</td>\n",
       "      <td>0.811370</td>\n",
       "      <td>2378</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>aerial</td>\n",
       "      <td>629</td>\n",
       "      <td>222</td>\n",
       "      <td>0.352941</td>\n",
       "      <td>1000</td>\n",
       "      <td>446</td>\n",
       "      <td>0.446</td>\n",
       "      <td>1000</td>\n",
       "      <td>22</td>\n",
       "      <td>0.864865</td>\n",
       "      <td>2546</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>romance</td>\n",
       "      <td>1000</td>\n",
       "      <td>161</td>\n",
       "      <td>0.161000</td>\n",
       "      <td>1000</td>\n",
       "      <td>337</td>\n",
       "      <td>0.337</td>\n",
       "      <td>1000</td>\n",
       "      <td>3</td>\n",
       "      <td>0.928000</td>\n",
       "      <td>2920</td>\n",
       "      <td>genres</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>interview</td>\n",
       "      <td>1000</td>\n",
       "      <td>230</td>\n",
       "      <td>0.230000</td>\n",
       "      <td>1000</td>\n",
       "      <td>236</td>\n",
       "      <td>0.236</td>\n",
       "      <td>1000</td>\n",
       "      <td>3</td>\n",
       "      <td>0.909000</td>\n",
       "      <td>2935</td>\n",
       "      <td>events_actions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>tilt-shot</td>\n",
       "      <td>1000</td>\n",
       "      <td>272</td>\n",
       "      <td>0.272000</td>\n",
       "      <td>1000</td>\n",
       "      <td>77</td>\n",
       "      <td>0.077</td>\n",
       "      <td>1000</td>\n",
       "      <td>68</td>\n",
       "      <td>0.793000</td>\n",
       "      <td>2975</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>dutch-angle</td>\n",
       "      <td>1000</td>\n",
       "      <td>115</td>\n",
       "      <td>0.115000</td>\n",
       "      <td>1000</td>\n",
       "      <td>73</td>\n",
       "      <td>0.073</td>\n",
       "      <td>1000</td>\n",
       "      <td>32</td>\n",
       "      <td>0.665000</td>\n",
       "      <td>2977</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>medium</td>\n",
       "      <td>1000</td>\n",
       "      <td>457</td>\n",
       "      <td>0.457000</td>\n",
       "      <td>1000</td>\n",
       "      <td>78</td>\n",
       "      <td>0.078</td>\n",
       "      <td>1000</td>\n",
       "      <td>282</td>\n",
       "      <td>0.782000</td>\n",
       "      <td>2991</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>nudity</td>\n",
       "      <td>1000</td>\n",
       "      <td>219</td>\n",
       "      <td>0.219000</td>\n",
       "      <td>1000</td>\n",
       "      <td>157</td>\n",
       "      <td>0.157</td>\n",
       "      <td>1000</td>\n",
       "      <td>1</td>\n",
       "      <td>0.639000</td>\n",
       "      <td>2857</td>\n",
       "      <td>sensitivities</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>wide</td>\n",
       "      <td>298</td>\n",
       "      <td>168</td>\n",
       "      <td>0.563758</td>\n",
       "      <td>1000</td>\n",
       "      <td>565</td>\n",
       "      <td>0.565</td>\n",
       "      <td>1000</td>\n",
       "      <td>179</td>\n",
       "      <td>0.889262</td>\n",
       "      <td>2280</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>gore</td>\n",
       "      <td>1001</td>\n",
       "      <td>326</td>\n",
       "      <td>0.325674</td>\n",
       "      <td>1000</td>\n",
       "      <td>810</td>\n",
       "      <td>0.810</td>\n",
       "      <td>1000</td>\n",
       "      <td>11</td>\n",
       "      <td>0.846154</td>\n",
       "      <td>2844</td>\n",
       "      <td>sensitivities</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>scared</td>\n",
       "      <td>1000</td>\n",
       "      <td>217</td>\n",
       "      <td>0.217000</td>\n",
       "      <td>1000</td>\n",
       "      <td>152</td>\n",
       "      <td>0.152</td>\n",
       "      <td>1000</td>\n",
       "      <td>6</td>\n",
       "      <td>0.719000</td>\n",
       "      <td>2962</td>\n",
       "      <td>emotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>shutter-shot</td>\n",
       "      <td>1000</td>\n",
       "      <td>19</td>\n",
       "      <td>0.019000</td>\n",
       "      <td>1000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.988000</td>\n",
       "      <td>2986</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>group-shot</td>\n",
       "      <td>300</td>\n",
       "      <td>157</td>\n",
       "      <td>0.523333</td>\n",
       "      <td>1000</td>\n",
       "      <td>847</td>\n",
       "      <td>0.847</td>\n",
       "      <td>1000</td>\n",
       "      <td>125</td>\n",
       "      <td>0.760000</td>\n",
       "      <td>2267</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>closeup</td>\n",
       "      <td>460</td>\n",
       "      <td>223</td>\n",
       "      <td>0.484783</td>\n",
       "      <td>1000</td>\n",
       "      <td>793</td>\n",
       "      <td>0.793</td>\n",
       "      <td>1000</td>\n",
       "      <td>188</td>\n",
       "      <td>0.950000</td>\n",
       "      <td>2409</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>extreme-wide-shot</td>\n",
       "      <td>1000</td>\n",
       "      <td>294</td>\n",
       "      <td>0.294000</td>\n",
       "      <td>1000</td>\n",
       "      <td>600</td>\n",
       "      <td>0.600</td>\n",
       "      <td>1000</td>\n",
       "      <td>27</td>\n",
       "      <td>0.854000</td>\n",
       "      <td>2928</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>run</td>\n",
       "      <td>1000</td>\n",
       "      <td>239</td>\n",
       "      <td>0.239000</td>\n",
       "      <td>1000</td>\n",
       "      <td>357</td>\n",
       "      <td>0.357</td>\n",
       "      <td>1000</td>\n",
       "      <td>8</td>\n",
       "      <td>0.882000</td>\n",
       "      <td>2909</td>\n",
       "      <td>events_actions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>interior</td>\n",
       "      <td>343</td>\n",
       "      <td>177</td>\n",
       "      <td>0.516035</td>\n",
       "      <td>1000</td>\n",
       "      <td>684</td>\n",
       "      <td>0.684</td>\n",
       "      <td>1000</td>\n",
       "      <td>597</td>\n",
       "      <td>0.845481</td>\n",
       "      <td>2312</td>\n",
       "      <td>time_location</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>horror</td>\n",
       "      <td>570</td>\n",
       "      <td>201</td>\n",
       "      <td>0.352632</td>\n",
       "      <td>1000</td>\n",
       "      <td>830</td>\n",
       "      <td>0.830</td>\n",
       "      <td>1000</td>\n",
       "      <td>45</td>\n",
       "      <td>0.847368</td>\n",
       "      <td>2492</td>\n",
       "      <td>genres</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>jump-scare</td>\n",
       "      <td>1000</td>\n",
       "      <td>9</td>\n",
       "      <td>0.009000</td>\n",
       "      <td>1000</td>\n",
       "      <td>4</td>\n",
       "      <td>0.004</td>\n",
       "      <td>1000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.955000</td>\n",
       "      <td>2930</td>\n",
       "      <td>motion</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>fantasy</td>\n",
       "      <td>1000</td>\n",
       "      <td>388</td>\n",
       "      <td>0.388000</td>\n",
       "      <td>1000</td>\n",
       "      <td>427</td>\n",
       "      <td>0.427</td>\n",
       "      <td>1000</td>\n",
       "      <td>25</td>\n",
       "      <td>0.816000</td>\n",
       "      <td>2936</td>\n",
       "      <td>genres</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>character-focus</td>\n",
       "      <td>303</td>\n",
       "      <td>187</td>\n",
       "      <td>0.617162</td>\n",
       "      <td>1000</td>\n",
       "      <td>904</td>\n",
       "      <td>0.904</td>\n",
       "      <td>1000</td>\n",
       "      <td>695</td>\n",
       "      <td>0.775578</td>\n",
       "      <td>2299</td>\n",
       "      <td>focus</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>sad</td>\n",
       "      <td>1019</td>\n",
       "      <td>305</td>\n",
       "      <td>0.299313</td>\n",
       "      <td>1000</td>\n",
       "      <td>326</td>\n",
       "      <td>0.326</td>\n",
       "      <td>1000</td>\n",
       "      <td>4</td>\n",
       "      <td>0.773307</td>\n",
       "      <td>2933</td>\n",
       "      <td>emotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>action</td>\n",
       "      <td>369</td>\n",
       "      <td>176</td>\n",
       "      <td>0.476965</td>\n",
       "      <td>1000</td>\n",
       "      <td>910</td>\n",
       "      <td>0.910</td>\n",
       "      <td>1000</td>\n",
       "      <td>201</td>\n",
       "      <td>0.837398</td>\n",
       "      <td>2346</td>\n",
       "      <td>genres</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>golden-hour</td>\n",
       "      <td>1000</td>\n",
       "      <td>85</td>\n",
       "      <td>0.085000</td>\n",
       "      <td>1000</td>\n",
       "      <td>265</td>\n",
       "      <td>0.265</td>\n",
       "      <td>1000</td>\n",
       "      <td>3</td>\n",
       "      <td>0.874000</td>\n",
       "      <td>2935</td>\n",
       "      <td>time_location</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>extreme-close-up</td>\n",
       "      <td>439</td>\n",
       "      <td>228</td>\n",
       "      <td>0.519362</td>\n",
       "      <td>1000</td>\n",
       "      <td>575</td>\n",
       "      <td>0.575</td>\n",
       "      <td>1000</td>\n",
       "      <td>31</td>\n",
       "      <td>0.917995</td>\n",
       "      <td>2396</td>\n",
       "      <td>shot_types</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     label   cnt  cnt_pos  pos_rate  cnt_zs  cnt_zs_pos  \\\n",
       "0                    happy  1000      287  0.287000    1000         689   \n",
       "1                     zoom  1000       66  0.066000    1000          66   \n",
       "2              static-shot  1000      451  0.451000    1000          16   \n",
       "3                low-angle   272      102  0.375000    1000         166   \n",
       "4                    drama   599      203  0.338898    1000         590   \n",
       "5                 violence   575      193  0.335652    1000         807   \n",
       "6              cowboy-shot  1000      373  0.373000    1000         121   \n",
       "7                   sci-fi   768      257  0.334635    1000         777   \n",
       "8                      day   249      124  0.497992    1000         505   \n",
       "9                 intimacy  1000      336  0.336000    1000         473   \n",
       "10               car-chase  1000      403  0.403000    1000         408   \n",
       "11      establishing-shots   938      449  0.478678    1000         870   \n",
       "12               timelapse  1003        6  0.005982    1000          12   \n",
       "13              high-angle   484      225  0.464876    1000         598   \n",
       "14                 smoking  1000      217  0.217000    1000         194   \n",
       "15           overhead-shot   513      132  0.257310    1000         342   \n",
       "16                laughter  1000      307  0.307000    1000         227   \n",
       "17               eye-level  1000      675  0.675000    1000         482   \n",
       "18                   drugs  1000      107  0.107000    1000         119   \n",
       "19                   fight  1000      285  0.285000    1000         715   \n",
       "20                 alcohol  1000      268  0.268000    1000         269   \n",
       "21                   anger  1000      215  0.215000    1000         257   \n",
       "22                two-shot   402      246  0.611940    1000          37   \n",
       "23                     pan  1000      154  0.154000    1000          28   \n",
       "24                  object   351      136  0.387464    1000         729   \n",
       "25                  animal   420      166  0.395238    1000         769   \n",
       "26             slow-motion  1000       64  0.064000    1000         117   \n",
       "27  over-the-shoulder-shot   637      295  0.463108    1000          16   \n",
       "28                handheld  1000      340  0.340000    1000          91   \n",
       "29      point-of-view-shot  1000      201  0.201000    1000          50   \n",
       "30             single-shot   357      159  0.445378    1000         117   \n",
       "31             insert-shot   387      194  0.501292    1000         178   \n",
       "32                  aerial   629      222  0.352941    1000         446   \n",
       "33                 romance  1000      161  0.161000    1000         337   \n",
       "34               interview  1000      230  0.230000    1000         236   \n",
       "35               tilt-shot  1000      272  0.272000    1000          77   \n",
       "36             dutch-angle  1000      115  0.115000    1000          73   \n",
       "37                  medium  1000      457  0.457000    1000          78   \n",
       "38                  nudity  1000      219  0.219000    1000         157   \n",
       "39                    wide   298      168  0.563758    1000         565   \n",
       "40                    gore  1001      326  0.325674    1000         810   \n",
       "41                  scared  1000      217  0.217000    1000         152   \n",
       "42            shutter-shot  1000       19  0.019000    1000           0   \n",
       "43              group-shot   300      157  0.523333    1000         847   \n",
       "44                 closeup   460      223  0.484783    1000         793   \n",
       "45       extreme-wide-shot  1000      294  0.294000    1000         600   \n",
       "46                     run  1000      239  0.239000    1000         357   \n",
       "47                interior   343      177  0.516035    1000         684   \n",
       "48                  horror   570      201  0.352632    1000         830   \n",
       "49              jump-scare  1000        9  0.009000    1000           4   \n",
       "50                 fantasy  1000      388  0.388000    1000         427   \n",
       "51         character-focus   303      187  0.617162    1000         904   \n",
       "52                     sad  1019      305  0.299313    1000         326   \n",
       "53                  action   369      176  0.476965    1000         910   \n",
       "54             golden-hour  1000       85  0.085000    1000         265   \n",
       "55        extreme-close-up   439      228  0.519362    1000         575   \n",
       "\n",
       "    zs_pos_rate  cnt_rand  cnt_rand_pos  agreement  unique_cnt           group  \n",
       "0         0.689      1000            33   0.843000        2927        emotions  \n",
       "1         0.066      1000            26   0.890000        2929          motion  \n",
       "2         0.016      1000            73   0.881000        2995      shot_types  \n",
       "3         0.166      1000            35   0.852941        2253      shot_types  \n",
       "4         0.590      1000           129   0.646077        2563          genres  \n",
       "5         0.807      1000           109   0.787826        2546   sensitivities  \n",
       "6         0.121      1000            17   0.848000        2934      shot_types  \n",
       "7         0.777      1000            21   0.842448        2712          genres  \n",
       "8         0.505      1000           370   0.975904        2233   time_location  \n",
       "9         0.473      1000            15   0.904000        2909   sensitivities  \n",
       "10        0.408      1000             8   0.778000        2921  events_actions  \n",
       "11        0.870      1000            57   0.772921        2863      shot_types  \n",
       "12        0.012      1000             0   0.987039        2944          motion  \n",
       "13        0.598      1000            38   0.845041        2430      shot_types  \n",
       "14        0.194      1000             4   0.953000        2844   sensitivities  \n",
       "15        0.342      1000             8   0.740741        2481      shot_types  \n",
       "16        0.227      1000             5   0.795000        2888        emotions  \n",
       "17        0.482      1000           555   0.642000        2986      shot_types  \n",
       "18        0.119      1000             5   0.871000        2860   sensitivities  \n",
       "19        0.715      1000            17   0.868000        2938  events_actions  \n",
       "20        0.269      1000             0   0.780000        2859   sensitivities  \n",
       "21        0.257      1000            21   0.624000        2876        emotions  \n",
       "22        0.037      1000            76   0.875622        2399      shot_types  \n",
       "23        0.028      1000            12   0.870000        2976          motion  \n",
       "24        0.729      1000            68   0.823362        2347           focus  \n",
       "25        0.769      1000            10   0.940476        2384           focus  \n",
       "26        0.117      1000             7   0.939000        2915          motion  \n",
       "27        0.016      1000            38   0.897959        2633      shot_types  \n",
       "28        0.091      1000            52   0.813000        2992          motion  \n",
       "29        0.050      1000             1   0.844000        2949      shot_types  \n",
       "30        0.117      1000           271   0.795518        2350      shot_types  \n",
       "31        0.178      1000            60   0.811370        2378      shot_types  \n",
       "32        0.446      1000            22   0.864865        2546      shot_types  \n",
       "33        0.337      1000             3   0.928000        2920          genres  \n",
       "34        0.236      1000             3   0.909000        2935  events_actions  \n",
       "35        0.077      1000            68   0.793000        2975      shot_types  \n",
       "36        0.073      1000            32   0.665000        2977      shot_types  \n",
       "37        0.078      1000           282   0.782000        2991      shot_types  \n",
       "38        0.157      1000             1   0.639000        2857   sensitivities  \n",
       "39        0.565      1000           179   0.889262        2280      shot_types  \n",
       "40        0.810      1000            11   0.846154        2844   sensitivities  \n",
       "41        0.152      1000             6   0.719000        2962        emotions  \n",
       "42        0.000      1000             0   0.988000        2986      shot_types  \n",
       "43        0.847      1000           125   0.760000        2267      shot_types  \n",
       "44        0.793      1000           188   0.950000        2409      shot_types  \n",
       "45        0.600      1000            27   0.854000        2928      shot_types  \n",
       "46        0.357      1000             8   0.882000        2909  events_actions  \n",
       "47        0.684      1000           597   0.845481        2312   time_location  \n",
       "48        0.830      1000            45   0.847368        2492          genres  \n",
       "49        0.004      1000             0   0.955000        2930          motion  \n",
       "50        0.427      1000            25   0.816000        2936          genres  \n",
       "51        0.904      1000           695   0.775578        2299           focus  \n",
       "52        0.326      1000             4   0.773307        2933        emotions  \n",
       "53        0.910      1000           201   0.837398        2346          genres  \n",
       "54        0.265      1000             3   0.874000        2935   time_location  \n",
       "55        0.575      1000            31   0.917995        2396      shot_types  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stats = pd.DataFrame(stats(l) for l in cfg.LABELS)\n",
    "df_stats.drop(columns=['keys', 'keys_unique'])"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "c4050f35-de40-40f4-92b8-444824b984c7",
   "metadata": {},
   "source": [
    "## label group aggregation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "73250ed1-e5d3-4437-8aaf-a39ab8a6af42",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:22.820930Z",
     "iopub.status.busy": "2024-02-02T22:36:22.820342Z",
     "iopub.status.idle": "2024-02-02T22:36:22.840795Z",
     "shell.execute_reply": "2024-02-02T22:36:22.840220Z",
     "shell.execute_reply.started": "2024-02-02T22:36:22.820902Z"
    },
    "tags": []
   },
   "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>label_cnt</th>\n",
       "      <th>va_cnt_mean</th>\n",
       "      <th>unique_cnt_mean</th>\n",
       "      <th>agreement_pct_mean</th>\n",
       "      <th>pos_pct_mean</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>group</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>emotions</th>\n",
       "      <td>5</td>\n",
       "      <td>1003</td>\n",
       "      <td>2917</td>\n",
       "      <td>75</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>events_actions</th>\n",
       "      <td>4</td>\n",
       "      <td>1000</td>\n",
       "      <td>2925</td>\n",
       "      <td>85</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>focus</th>\n",
       "      <td>3</td>\n",
       "      <td>358</td>\n",
       "      <td>2343</td>\n",
       "      <td>84</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>genres</th>\n",
       "      <td>6</td>\n",
       "      <td>717</td>\n",
       "      <td>2661</td>\n",
       "      <td>82</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>motion</th>\n",
       "      <td>6</td>\n",
       "      <td>1000</td>\n",
       "      <td>2947</td>\n",
       "      <td>90</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sensitivities</th>\n",
       "      <td>7</td>\n",
       "      <td>939</td>\n",
       "      <td>2817</td>\n",
       "      <td>82</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>shot_types</th>\n",
       "      <td>22</td>\n",
       "      <td>687</td>\n",
       "      <td>2654</td>\n",
       "      <td>83</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time_location</th>\n",
       "      <td>3</td>\n",
       "      <td>530</td>\n",
       "      <td>2493</td>\n",
       "      <td>89</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                label_cnt  va_cnt_mean  unique_cnt_mean  agreement_pct_mean  \\\n",
       "group                                                                         \n",
       "emotions                5         1003             2917                  75   \n",
       "events_actions          4         1000             2925                  85   \n",
       "focus                   3          358             2343                  84   \n",
       "genres                  6          717             2661                  82   \n",
       "motion                  6         1000             2947                  90   \n",
       "sensitivities           7          939             2817                  82   \n",
       "shot_types             22          687             2654                  83   \n",
       "time_location           3          530             2493                  89   \n",
       "\n",
       "                pos_pct_mean  \n",
       "group                         \n",
       "emotions                  26  \n",
       "events_actions            28  \n",
       "focus                     46  \n",
       "genres                    34  \n",
       "motion                    10  \n",
       "sensitivities             25  \n",
       "shot_types                40  \n",
       "time_location             36  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stats.groupby('group').apply(\n",
    "    lambda x: pd.Series(dict(\n",
    "        label_cnt=len(x),\n",
    "        va_cnt_mean=np.mean(x.cnt),\n",
    "        unique_cnt_mean=np.mean(x.unique_cnt),\n",
    "        agreement_pct_mean=np.mean(x.agreement) * 100,\n",
    "        pos_pct_mean=np.mean(x.pos_rate) * 100,\n",
    "    ))\n",
    ").round(1).astype(int)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "a4058742-e2bc-464a-b98f-0ff1e6ecaa3a",
   "metadata": {},
   "source": [
    "### visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "01aefd75-c923-46fa-9f9d-afeae58fdca1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:24.069330Z",
     "iopub.status.busy": "2024-02-02T22:36:24.068633Z",
     "iopub.status.idle": "2024-02-02T22:36:25.842729Z",
     "shell.execute_reply": "2024-02-02T22:36:25.841996Z",
     "shell.execute_reply.started": "2024-02-02T22:36:24.069299Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def get_row(l):\n",
    "    ds = data.get_aggregate_labeled_dataset(label=l)\n",
    "    return dict(\n",
    "        label=l,\n",
    "        pos_rate=ds.pos_rate,\n",
    "        pos_cnt=ds.pos_cnt,\n",
    "        data_score=ds.data_score,\n",
    "        cnt=len(ds),\n",
    "        group=l2g[l],\n",
    "    )\n",
    "\n",
    "d = pd.DataFrame(map(get_row, cfg.LABELS)).sort_values(by='pos_cnt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f0eb8e0a-94e4-434c-916b-6a09efcec277",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:25.844346Z",
     "iopub.status.busy": "2024-02-02T22:36:25.844022Z",
     "iopub.status.idle": "2024-02-02T22:36:25.848289Z",
     "shell.execute_reply": "2024-02-02T22:36:25.847715Z",
     "shell.execute_reply.started": "2024-02-02T22:36:25.844321Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def make_box_plot(col, name, xlim):\n",
    "    d.boxplot(col, by='group', vert=False)\n",
    "    plt.xlim(xlim)\n",
    "    plt.suptitle('')\n",
    "    plt.xlabel('')\n",
    "    _ = plt.title(f'Distribution of of {name} by label group')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "a52debcb-6f3b-4ab8-afd5-c4d9e0445321",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:25.849507Z",
     "iopub.status.busy": "2024-02-02T22:36:25.849060Z",
     "iopub.status.idle": "2024-02-02T22:36:26.335290Z",
     "shell.execute_reply": "2024-02-02T22:36:26.334670Z",
     "shell.execute_reply.started": "2024-02-02T22:36:25.849484Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "make_box_plot(col='cnt', name='VA annotation counts', xlim=[0, 1_020])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "9218ebaa-cfcf-4874-8176-c0355a1a10eb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:26.337107Z",
     "iopub.status.busy": "2024-02-02T22:36:26.336629Z",
     "iopub.status.idle": "2024-02-02T22:36:26.550041Z",
     "shell.execute_reply": "2024-02-02T22:36:26.549453Z",
     "shell.execute_reply.started": "2024-02-02T22:36:26.337082Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "make_box_plot(col='pos_rate', name='positive rate', xlim=[0, 1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2e89cfc7-dad2-4352-a129-e6433e18358d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:26.551361Z",
     "iopub.status.busy": "2024-02-02T22:36:26.550893Z",
     "iopub.status.idle": "2024-02-02T22:36:26.799561Z",
     "shell.execute_reply": "2024-02-02T22:36:26.798967Z",
     "shell.execute_reply.started": "2024-02-02T22:36:26.551336Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "make_box_plot(col='data_score', name='data score', xlim=[0, 1])"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "6da54083-65e9-4216-8eb6-b525249f2540",
   "metadata": {},
   "source": [
    "## global stats across all labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "041b8897-81c2-4678-8784-b70bdb79b952",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:27.408834Z",
     "iopub.status.busy": "2024-02-02T22:36:27.408216Z",
     "iopub.status.idle": "2024-02-02T22:36:27.485319Z",
     "shell.execute_reply": "2024-02-02T22:36:27.484729Z",
     "shell.execute_reply.started": "2024-02-02T22:36:27.408806Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "70852"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# number of unique clips annotated\n",
    "len(functools.reduce(set.union, df_stats.keys_unique))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "438a9e9e-ddba-435d-9f06-2e889aa13202",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:27.794424Z",
     "iopub.status.busy": "2024-02-02T22:36:27.794052Z",
     "iopub.status.idle": "2024-02-02T22:36:27.798654Z",
     "shell.execute_reply": "2024-02-02T22:36:27.798079Z",
     "shell.execute_reply.started": "2024-02-02T22:36:27.794399Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "43686"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# number of VA annotations\n",
    "df_stats.cnt.sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "3dcd44d2-937c-4b9a-b68d-1a5a2736c745",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:28.274419Z",
     "iopub.status.busy": "2024-02-02T22:36:28.273996Z",
     "iopub.status.idle": "2024-02-02T22:36:28.279688Z",
     "shell.execute_reply": "2024-02-02T22:36:28.278959Z",
     "shell.execute_reply.started": "2024-02-02T22:36:28.274391Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "152579"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# number of total annotations (VA + zero shot + random)\n",
    "df_stats.unique_cnt.sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d4760ec7-e7f5-4065-bd18-e558664cca2e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:28.717162Z",
     "iopub.status.busy": "2024-02-02T22:36:28.716449Z",
     "iopub.status.idle": "2024-02-02T22:36:28.724285Z",
     "shell.execute_reply": "2024-02-02T22:36:28.723716Z",
     "shell.execute_reply.started": "2024-02-02T22:36:28.717132Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    56.000000\n",
       "mean      0.836601\n",
       "std       0.087457\n",
       "min       0.624000\n",
       "25%       0.786370\n",
       "50%       0.845817\n",
       "75%       0.889446\n",
       "max       0.988000\n",
       "Name: agreement, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# agreement distribution across all labels\n",
    "df_stats.agreement.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "5303fdc9-e594-47bd-bbed-918bafb0b4be",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-02-02T22:36:29.210924Z",
     "iopub.status.busy": "2024-02-02T22:36:29.210270Z",
     "iopub.status.idle": "2024-02-02T22:36:29.217866Z",
     "shell.execute_reply": "2024-02-02T22:36:29.217263Z",
     "shell.execute_reply.started": "2024-02-02T22:36:29.210896Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    56.000000\n",
       "mean      0.328204\n",
       "std       0.161572\n",
       "min       0.005982\n",
       "25%       0.218500\n",
       "50%       0.335826\n",
       "75%       0.458527\n",
       "max       0.675000\n",
       "Name: pos_rate, dtype: float64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# positive rate across all labels\n",
    "df_stats.pos_rate.describe()"
   ]
  }
 ],
 "metadata": {
  "hide_input": false,
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}