hbredin/generic.ipynb

## generic.ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "%pylab inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyannote.database import Database\n",
    "import pyannote.database.protocol\n",
    "import functools\n",
    "import pandas as pd\n",
    "from pyannote.core import Segment, Timeline, Annotation\n",
    "from pathlib import Path\n",
    "import yaml"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "REPERE:\r\n",
      "  SpeakerDiarization:\r\n",
      "    Phase2:\r\n",
      "      train:\r\n",
      "        annotation: /Users/bredin/Development/pyannote/pyannote-db/pyannote-db-repere/REPERE/data/speaker/phase2.trn.mdtm\r\n",
      "        annotated: /Users/bredin/Development/pyannote/pyannote-db/pyannote-db-repere/REPERE/data/speaker/phase2.trn.uem\r\n",
      "      development:\r\n",
      "        annotation: /Users/bredin/Development/pyannote/pyannote-db/pyannote-db-repere/REPERE/data/speaker/phase2.dev.mdtm\r\n",
      "        annotated: /Users/bredin/Development/pyannote/pyannote-db/pyannote-db-repere/REPERE/data/speaker/phase2.dev.uem\r\n",
      "      test:\r\n",
      "        annotation: /Users/bredin/Development/pyannote/pyannote-db/pyannote-db-repere/REPERE/data/speaker/phase2.tst.mdtm\r\n",
      "        annotated: /Users/bredin/Development/pyannote/pyannote-db/pyannote-db-repere/REPERE/data/speaker/phase2.tst.uem\r\n"
     ]
    }
   ],
   "source": [
    "!cat ~/.pyannote/protocols.yml"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = Path('~/.pyannote/protocols.yml').expanduser()\n",
    "with open(path, 'r') as fp:\n",
    "    content = yaml.load(fp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def subset_iter(database_name, path_mdtm=None, path_uem=None):\n",
    "    \"\"\"This function will become a xxx_iter method of a protocol.\n",
    "    \n",
    "    Parameters\n",
    "    ----------\n",
    "    database_name : `str`\n",
    "        Database name.\n",
    "    path_mdtm : `Path`, optional\n",
    "        Path to MDTM file.\n",
    "    path_uem : `Path`, optional\n",
    "        Path to UEM file.\n",
    "    \n",
    "    Yields\n",
    "    ------\n",
    "    current_file : `dict`\n",
    "        Dictionary that provides information about a file.\n",
    "        This includes (at least) a 'uri' key and a 'database' keys.\n",
    "    \n",
    "    \"\"\"\n",
    "    \n",
    "    names = ['uri', 'NA1', 'start', 'duration', 'NA2', 'NA3', 'NA4', 'label']\n",
    "    annotations = pd.read_table(path_mdtm, names=names, delim_whitespace=True)\n",
    "    \n",
    "    names = ['uri', 'NA1', 'start', 'end']\n",
    "    annotated = pd.read_table(path_uem, names=names, delim_whitespace=True).groupby('uri')\n",
    "        \n",
    "    for uri, turns in annotations.groupby('uri'):\n",
    "        reference = Annotation(uri=uri)\n",
    "        for turn_id, row in turns.iterrows():\n",
    "            segment = Segment(row.start, row.start + row.duration)\n",
    "            label = row.label\n",
    "            reference[segment, turn_id] = label\n",
    "            \n",
    "        uem = Timeline(uri=uri)\n",
    "        for _, row in annotated.get_group(uri).iterrows():\n",
    "            segment = Segment(row.start, row.end)\n",
    "            uem.add(segment)\n",
    "        \n",
    "        yield {'uri': uri,\n",
    "               'database': database_name,\n",
    "               'annotation': reference,\n",
    "               'annotated': uem}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# this dictionary is meant to contain one class per main entry in ~/.pyannote/protocols.yml\n",
    "databases = {}\n",
    "\n",
    "# for each database\n",
    "for database_name, tasks in content.items():\n",
    "\n",
    "    # this list is meant to contain one class per protocol\n",
    "    register = []\n",
    "    \n",
    "    # for each type of task\n",
    "    for task_name, protocols in tasks.items():\n",
    "\n",
    "        # only speaker diarization tasks are supported for now...\n",
    "        if task_name != 'SpeakerDiarization':\n",
    "            msg = ('Only speaker diarization protocols are supported for now.')\n",
    "            raise ValueError(msg)\n",
    "            \n",
    "\n",
    "        # get protocol base class (here: SpeakerDiarizationProtocol)\n",
    "        protocol_base_class = getattr(pyannote.database.protocol, \n",
    "                                      f'{task_name}Protocol')\n",
    "\n",
    "        # annoying mapping...\n",
    "        SUBSET = {'train': 'trn', 'development': 'dev', 'test': 'tst'}\n",
    "\n",
    "        # for each protocol\n",
    "        for protocol_name, subsets in protocols.items():\n",
    "            \n",
    "            # this dictionary is meant to contain \"trn_iter\", \"dev_iter\", and \"tst_iter\" methods\n",
    "            protocol_methods = {}\n",
    "            \n",
    "            # for each subset\n",
    "            for subset, sub in SUBSET.items():\n",
    "                \n",
    "                # if protocols.yml does not provide a file for this subset, skip it\n",
    "                if subset not in subsets:\n",
    "                    continue\n",
    "                \n",
    "                # define xxx_iter method\n",
    "                protocol_methods[f'{sub}_iter'] = functools.partial(\n",
    "                    subset_iter, \n",
    "                    database_name,\n",
    "                    path_mdtm=subsets[subset].get('annotation', None),\n",
    "                    path_uem=subsets[subset].get('annotated', None))\n",
    "\n",
    "        # create protocol class on-the-fly\n",
    "        protocol = type(protocol_name, (protocol_base_class,), protocol_methods)\n",
    "        \n",
    "        # keep track of this protocol -- database.__init__ needs to register it later...\n",
    "        register.append((task_name, protocol_name, protocol))\n",
    "    \n",
    "    # define Database.__init__ method\n",
    "    def db_init(self, preprocessors={}):\n",
    "        super(self.__class__, self).__init__(preprocessors=preprocessors)\n",
    "        for protocol in register:\n",
    "            self.register_protocol(*protocol)\n",
    "    \n",
    "    # create database class on-the-fly\n",
    "    databases[database_name] = type(database_name, (Database,), {'__init__': db_init})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "preprocessors = {}\n",
    "db = databases['REPERE'](preprocessors=preprocessors)\n",
    "protocol = db.get_protocol('SpeakerDiarization', 'Phase2')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "for current_file in protocol.test():\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'uri': 'BFMTV_BFMStory_2012-07-24_175800',\n",
       " 'database': 'REPERE',\n",
       " 'annotation': <pyannote.core.annotation.Annotation at 0x3103a1c88>,\n",
       " 'annotated': <Timeline(uri=BFMTV_BFMStory_2012-07-24_175800, segments=[<Segment(53.868, 3374.15)>])>}"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "current_file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Timeline(uri=BFMTV_BFMStory_2012-07-24_175800, segments=[<Segment(53.868, 3374.15)>])>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "current_file['annotated']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<pyannote.core.annotation.Annotation at 0x3103a1c88>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "current_file['annotation']"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}