notebooks/B1_subset_one_file.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The goal of this notebook will be to find regions that max/minimizes a criterion in an audio file.\n",
    "\n",
    "For example we have a 1h conversation and we want to find what 20% of the data maximizes some heuristic based on the model output probabilities (e.g. lowest confidence).\n",
    "\n",
    "In practice we would most likely want to perform this sampling for the whole dataset rather than filewise (i.e. find the aforementioned 20% in the whole dataset rather than find it for each individual file). But let's keep the scope smaller and make things easier for ourselves !"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Loading the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "PROTOCOL=\"MSDWILD.SpeakerDiarization.OriginalMany\"\n",
    "PROTOCOL_SUBSET='test'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First let's retrieve the data we previously generated into DataFrame `infdf` aswell as its associated metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>uem</th>\n",
       "      <th>out_0</th>\n",
       "      <th>out_1</th>\n",
       "      <th>out_2</th>\n",
       "      <th>out_3</th>\n",
       "      <th>out_4</th>\n",
       "      <th>out_5</th>\n",
       "      <th>out_6</th>\n",
       "      <th>ref_0</th>\n",
       "      <th>ref_1</th>\n",
       "      <th>ref_2</th>\n",
       "      <th>ref_3</th>\n",
       "      <th>ref_4</th>\n",
       "      <th>ref_5</th>\n",
       "      <th>dataset</th>\n",
       "      <th>uri</th>\n",
       "      <th>ref_6</th>\n",
       "      <th>ref_7</th>\n",
       "      <th>ref_8</th>\n",
       "      <th>ref_9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>True</td>\n",
       "      <td>0.009554</td>\n",
       "      <td>0.008599</td>\n",
       "      <td>0.749115</td>\n",
       "      <td>0.202438</td>\n",
       "      <td>0.002572</td>\n",
       "      <td>0.001593</td>\n",
       "      <td>0.026128</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>MSDWILD</td>\n",
       "      <td>00004</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>True</td>\n",
       "      <td>0.007836</td>\n",
       "      <td>0.007699</td>\n",
       "      <td>0.756682</td>\n",
       "      <td>0.201422</td>\n",
       "      <td>0.002065</td>\n",
       "      <td>0.001207</td>\n",
       "      <td>0.023090</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>MSDWILD</td>\n",
       "      <td>00004</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>True</td>\n",
       "      <td>0.007783</td>\n",
       "      <td>0.007697</td>\n",
       "      <td>0.760542</td>\n",
       "      <td>0.199639</td>\n",
       "      <td>0.001927</td>\n",
       "      <td>0.001188</td>\n",
       "      <td>0.021223</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>MSDWILD</td>\n",
       "      <td>00004</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>True</td>\n",
       "      <td>0.007709</td>\n",
       "      <td>0.007748</td>\n",
       "      <td>0.763758</td>\n",
       "      <td>0.199777</td>\n",
       "      <td>0.001693</td>\n",
       "      <td>0.001086</td>\n",
       "      <td>0.018230</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>MSDWILD</td>\n",
       "      <td>00004</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>True</td>\n",
       "      <td>0.007464</td>\n",
       "      <td>0.007655</td>\n",
       "      <td>0.767000</td>\n",
       "      <td>0.199589</td>\n",
       "      <td>0.001482</td>\n",
       "      <td>0.000994</td>\n",
       "      <td>0.015816</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>MSDWILD</td>\n",
       "      <td>00004</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</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",
       "      <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>1687480</th>\n",
       "      <td>False</td>\n",
       "      <td>0.973313</td>\n",
       "      <td>0.002382</td>\n",
       "      <td>0.017132</td>\n",
       "      <td>0.006826</td>\n",
       "      <td>0.000092</td>\n",
       "      <td>0.000035</td>\n",
       "      <td>0.000219</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>MSDWILD</td>\n",
       "      <td>03141</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1687481</th>\n",
       "      <td>False</td>\n",
       "      <td>0.970997</td>\n",
       "      <td>0.002619</td>\n",
       "      <td>0.018684</td>\n",
       "      <td>0.007319</td>\n",
       "      <td>0.000103</td>\n",
       "      <td>0.000038</td>\n",
       "      <td>0.000242</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>MSDWILD</td>\n",
       "      <td>03141</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1687482</th>\n",
       "      <td>False</td>\n",
       "      <td>0.969682</td>\n",
       "      <td>0.002770</td>\n",
       "      <td>0.019347</td>\n",
       "      <td>0.007777</td>\n",
       "      <td>0.000114</td>\n",
       "      <td>0.000039</td>\n",
       "      <td>0.000270</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>MSDWILD</td>\n",
       "      <td>03141</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1687483</th>\n",
       "      <td>False</td>\n",
       "      <td>0.970696</td>\n",
       "      <td>0.002947</td>\n",
       "      <td>0.018097</td>\n",
       "      <td>0.007823</td>\n",
       "      <td>0.000125</td>\n",
       "      <td>0.000038</td>\n",
       "      <td>0.000273</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>MSDWILD</td>\n",
       "      <td>03141</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1687484</th>\n",
       "      <td>False</td>\n",
       "      <td>0.954898</td>\n",
       "      <td>0.005279</td>\n",
       "      <td>0.028204</td>\n",
       "      <td>0.010860</td>\n",
       "      <td>0.000251</td>\n",
       "      <td>0.000072</td>\n",
       "      <td>0.000436</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>MSDWILD</td>\n",
       "      <td>03141</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1687485 rows × 20 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           uem     out_0     out_1     out_2     out_3     out_4     out_5  \\\n",
       "0         True  0.009554  0.008599  0.749115  0.202438  0.002572  0.001593   \n",
       "1         True  0.007836  0.007699  0.756682  0.201422  0.002065  0.001207   \n",
       "2         True  0.007783  0.007697  0.760542  0.199639  0.001927  0.001188   \n",
       "3         True  0.007709  0.007748  0.763758  0.199777  0.001693  0.001086   \n",
       "4         True  0.007464  0.007655  0.767000  0.199589  0.001482  0.000994   \n",
       "...        ...       ...       ...       ...       ...       ...       ...   \n",
       "1687480  False  0.973313  0.002382  0.017132  0.006826  0.000092  0.000035   \n",
       "1687481  False  0.970997  0.002619  0.018684  0.007319  0.000103  0.000038   \n",
       "1687482  False  0.969682  0.002770  0.019347  0.007777  0.000114  0.000039   \n",
       "1687483  False  0.970696  0.002947  0.018097  0.007823  0.000125  0.000038   \n",
       "1687484  False  0.954898  0.005279  0.028204  0.010860  0.000251  0.000072   \n",
       "\n",
       "            out_6  ref_0  ref_1  ref_2  ref_3  ref_4  ref_5  dataset    uri  \\\n",
       "0        0.026128    0.0    0.0    0.0    0.0    0.0    0.0  MSDWILD  00004   \n",
       "1        0.023090    0.0    0.0    0.0    0.0    0.0    0.0  MSDWILD  00004   \n",
       "2        0.021223    0.0    1.0    0.0    0.0    0.0    0.0  MSDWILD  00004   \n",
       "3        0.018230    0.0    1.0    0.0    0.0    0.0    0.0  MSDWILD  00004   \n",
       "4        0.015816    0.0    1.0    0.0    0.0    0.0    0.0  MSDWILD  00004   \n",
       "...           ...    ...    ...    ...    ...    ...    ...      ...    ...   \n",
       "1687480  0.000219    0.0    0.0    0.0    0.0    0.0   -1.0  MSDWILD  03141   \n",
       "1687481  0.000242    0.0    0.0    0.0    0.0    0.0   -1.0  MSDWILD  03141   \n",
       "1687482  0.000270    0.0    0.0    0.0    0.0    0.0   -1.0  MSDWILD  03141   \n",
       "1687483  0.000273    0.0    0.0    0.0    0.0    0.0   -1.0  MSDWILD  03141   \n",
       "1687484  0.000436    0.0    0.0    0.0    0.0    0.0   -1.0  MSDWILD  03141   \n",
       "\n",
       "         ref_6  ref_7  ref_8  ref_9  \n",
       "0         -1.0   -1.0   -1.0   -1.0  \n",
       "1         -1.0   -1.0   -1.0   -1.0  \n",
       "2         -1.0   -1.0   -1.0   -1.0  \n",
       "3         -1.0   -1.0   -1.0   -1.0  \n",
       "4         -1.0   -1.0   -1.0   -1.0  \n",
       "...        ...    ...    ...    ...  \n",
       "1687480   -1.0   -1.0   -1.0   -1.0  \n",
       "1687481   -1.0   -1.0   -1.0   -1.0  \n",
       "1687482   -1.0   -1.0   -1.0   -1.0  \n",
       "1687483   -1.0   -1.0   -1.0   -1.0  \n",
       "1687484   -1.0   -1.0   -1.0   -1.0  \n",
       "\n",
       "[1687485 rows x 20 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from pathlib import Path\n",
    "from powerset_calibration.inference import get_inference_fid, load_inference_file, load_inference_metadata, InferenceMetadata\n",
    "\n",
    "fid = get_inference_fid('segmentation', PROTOCOL, PROTOCOL_SUBSET)\n",
    "fid_path = Path('outputs')/fid\n",
    "\n",
    "metadata: InferenceMetadata = load_inference_metadata(fid_path)\n",
    "infdf: pd.DataFrame = load_inference_file(fid_path)\n",
    "\n",
    "infdf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Preparing the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As a reminder, manually doing this stuff can be a bit hasardous, so for most uses I think it's easier to use `powerset_calibration.protocol.ActiveLearningProtocol` (or extend it/copy the code).\n",
    "\n",
    "But let's do things manually here.\n",
    "First let's convert our 'Inference Dataframe' data back to tensors. We will focus on one URI only."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "uri 03141\n",
      "uem torch.Size([12, 589, 1])\n",
      "ref torch.Size([12, 589, 5])\n",
      "out torch.Size([12, 589, 7])\n"
     ]
    }
   ],
   "source": [
    "from powerset_calibration.protocol import ActiveLearningProtocol\n",
    "import torch\n",
    "\n",
    "TREATED_URI = '03141' # set this to the URI you want to extract active learning windows on.\n",
    "\n",
    "infdf2 = infdf[infdf.uri == TREATED_URI]\n",
    "\n",
    "data = next(ActiveLearningProtocol.iterate_df_files(infdf2, metadata))\n",
    "\n",
    "for k,v in data.items():\n",
    "    if isinstance(v, torch.Tensor):\n",
    "        print(k, v.shape)\n",
    "    else:\n",
    "        print(k, v)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that tensors are of shape `(windows_count, num_frames, x)`, which means the inference data is **unaggregated**, that is to say we still have *overlapping* data from applying a sliding window with overlap.\n",
    "\n",
    "Pyannote inference (and the inference function from the first notebook) will aggregate things if they can be. But here we're dealing with the Speaker Diarization task, where any permutation of speaker identity is valid, meaning two classes might not mean the same things in two different outputs, so we cant aggregate them by just averaging their values (that's why there is an embedding extraction+clustering step in the pyannote pipeline).\n",
    "\n",
    "Before we do this aggregation let's set up some information we need."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Information about the model classes'"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from pyannote.core import SlidingWindow\n",
    "from pyannote.audio.utils.powerset import Powerset\n",
    "\n",
    "model_fps = metadata[\"model\"][\"fps\"]\n",
    "\n",
    "model_sw = SlidingWindow(metadata[\"model\"][\"duration\"], metadata[\"step_duration\"])\n",
    "\"\"\"This is the sliding window used to generate the inference file.\"\"\"\n",
    "\n",
    "model_frames = SlidingWindow(\n",
    "    metadata[\"model\"][\"frames\"][\"duration\"],\n",
    "    metadata[\"model\"][\"frames\"][\"step\"],\n",
    ")\n",
    "\"\"\"This precisely describes the alignment of the frames to the audio signal = receptive field\"\"\"\n",
    "\n",
    "powerset = Powerset(\n",
    "    len(metadata[\"model\"][\"specifications\"][\"classes\"]),\n",
    "    metadata[\"model\"][\"specifications\"][\"powerset_max_classes\"],\n",
    ")\n",
    "\"\"\"Information about the model classes\"\"\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can aggregate `uem` (easy), aswell as `out` and `ref` (harder).\n",
    "\n",
    "For prediction and reference, we ARE in (powerset) speaker diarization class space so we do have to use a smarter strategy. Here we rely on the 'SoftSegmentationPipeline' (using `aggregate_soft_segmentation`) which matches class spaces between windows using only their overlapped parts. It clearly couldn't be used for real diarization as their no notion of identity (persistence). But since we do not care about that (we want to retrieve model confidence/entropy/etc) it's all OK !\n",
    "\n",
    "(and it doesn't need any embedding/clustering, so it's really cheap)\n",
    "\n",
    "\n",
    "**Note**: Depending on the heuristic, we can actually \"aggregate late\". Here are the two ways to do aggregation (that we've thought of):\n",
    "- **Early aggregation**: preds/ref *(W,F,C)*  --agg-->  preds/ref  *(F2,C)*  -->  heuristic  *(F2,)*\n",
    "- **Late aggregation**: preds/ref *(W,F,C)* --> heuristic *(W,F,1)* --agg--> heuristics *(W,F,)*\n",
    "The advantage of late aggregation is that we do not need any fancy aggregation strategy since we're already down to a tensor with 1 feature, so no ambiguity."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.Size([3829]), torch.Size([3829, 7]), torch.Size([3829, 5]))"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from powerset_calibration.utils.inference import aggregate_sliding_window_tensor, aggregate_soft_segmentation\n",
    "\n",
    "# for UEM the aggregation doesnt exist as it only has one class (so we CAN just average the values, easy !)\n",
    "uem = aggregate_sliding_window_tensor(\n",
    "    tensor=data['uem'],\n",
    "    sliding_window=model_sw,\n",
    "    frames=model_frames,\n",
    ")[...,0].bool()\n",
    "\n",
    "pred = data['out']\n",
    "ref = data['ref']\n",
    "\n",
    "# Apply the soft segmentation-based aggregation\n",
    "pred_agg = aggregate_soft_segmentation(\n",
    "    t=data['out'],\n",
    "    sliding_window=model_sw,\n",
    "    powerset=powerset,\n",
    "    frames=model_frames,\n",
    ")\n",
    "ref_agg = aggregate_soft_segmentation(\n",
    "    t=data['ref'],\n",
    "    sliding_window=model_sw,\n",
    "    powerset=None,\n",
    "    frames=model_frames,\n",
    ")\n",
    "\n",
    "uem.shape, pred_agg.shape, ref_agg.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Perfect, now we have tensors of size `(num_frames2, X)` that cover the whole file !\n",
    "\n",
    "Now we have to decide what heuristic will be used to select high uncertainty region (or whatever we want to select). "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Computing relevant regions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.Size([3829]),\n",
       " torch.Size([3829]),\n",
       " torch.Size([3829]),\n",
       " torch.Size([3829]),\n",
       " torch.Size([3829]))"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from powerset_calibration.utils.inference import get_heuristic_tensor_segmentation\n",
    "\n",
    "# Just a shorthand to make things shorter\n",
    "def get_heur(heurtype, preds):\n",
    "    return get_heuristic_tensor_segmentation(\n",
    "        heuristic=heurtype,\n",
    "        preds=preds,\n",
    "        targets=ref,\n",
    "        sliding_window=model_sw,\n",
    "        frames=model_frames,\n",
    "        powerset=powerset,\n",
    "    )\n",
    "\n",
    "heur_rnd = get_heur('random', pred_agg)\n",
    "heur_conf = get_heur('confidence', pred_agg)\n",
    "heur_entropy = get_heur('entropy', pred_agg)\n",
    "\n",
    "heur_conf_lateagg = get_heur('confidence', pred)\n",
    "heur_entropy_lateagg = get_heur('entropy', pred)\n",
    "\n",
    "heur_rnd.shape, heur_conf.shape, heur_entropy.shape, heur_conf_lateagg.shape, heur_entropy_lateagg.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, let's find the relevant time regions according to our metrics."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from powerset_calibration.region_selector import generate_windows\n",
    "import sys, math\n",
    "\n",
    "SELECTED_REGIONS_MINSIZE = 5.0\n",
    "REGION_SEARCH_STEP = 0.5\n",
    "ANNOTATED_RATIO = 0.3\n",
    "\n",
    "OVERLAP_PENALITY = None\n",
    "REGIONS_PER_ITERATION = sys.maxsize\n",
    "\n",
    "# Uncomment this to try disabling any possible overlap between regions (ie they will be strictly SELECTED_REGIONS_MINSIZE long)\n",
    "# OVERLAP_PENALITY = math.inf\n",
    "# REGIONS_PER_ITERATION = 1\n",
    "\n",
    "def gen_wins(heur):\n",
    "    return generate_windows(\n",
    "        values=heur,\n",
    "        fps=model_fps,\n",
    "        window_duration=SELECTED_REGIONS_MINSIZE,\n",
    "        sliding_window_step=REGION_SEARCH_STEP,\n",
    "        annotated_ratio=ANNOTATED_RATIO,\n",
    "        penality=OVERLAP_PENALITY,\n",
    "        selections_per_iteration=REGIONS_PER_ITERATION,\n",
    "        uem=uem,\n",
    "    )\n",
    "\n",
    "selected_rnd = gen_wins(heur_rnd)\n",
    "selected_conf = gen_wins(heur_conf)\n",
    "selected_entropy = gen_wins(heur_entropy)\n",
    "selected_conf_lateagg = gen_wins(heur_conf_lateagg)\n",
    "selected_entropy_lateagg = gen_wins(heur_entropy_lateagg)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can visualize everything !\n",
    "For 'selected from X', the first line is the value of the heuristic at each point in time, and the second line shows the region that it ended up selecting.\n",
    "\n",
    "You can edit the parameter in the previous block to see different results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x1600 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def plot_tensor(ax, t,  title=None):\n",
    "    if isinstance(t, list):\n",
    "        t = [t2.squeeze()[...,None] for t2 in t]\n",
    "        t = torch.cat(t, dim=1)\n",
    "    ax.imshow(t.T, aspect='auto', interpolation='nearest')\n",
    "    ax.set_title(title)\n",
    "\n",
    "nrows = 8\n",
    "fig, axs = plt.subplots(nrows, 1, figsize=(15, nrows*2))\n",
    "plot_tensor(axs[0], uem[...,None], title='UEM')\n",
    "plot_tensor(axs[1], pred_agg, title='Pred aggregated')\n",
    "plot_tensor(axs[2], ref_agg, title='Ref aggregated')\n",
    "plot_tensor(axs[3], [heur_rnd, selected_rnd], title='Selected from \"random\"')\n",
    "plot_tensor(axs[4], [heur_conf, selected_conf], title='Selected from \"confidence\"')\n",
    "plot_tensor(axs[5], [heur_conf_lateagg, selected_conf_lateagg], title='Selected from \"confidence\" (late aggregation)')\n",
    "plot_tensor(axs[6], [heur_entropy, selected_entropy], title='Selected from \"entropy\"')\n",
    "plot_tensor(axs[7], [heur_entropy_lateagg, selected_entropy_lateagg], title='Selected from \"entropy\" (late aggregation)')\n",
    "\n",
    "\n",
    "fig.tight_layout()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And conversion from Tensor to Timeline is straightforward"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Timeline(uri=None, segments=[<Segment(5.90832, 11.9015)>, <Segment(29.5416, 34.5501)>, <Segment(41.8506, 47.3514)>, <Segment(57.6061, 59.3209)>])>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from powerset_calibration.utils.pyannote_core import tensor_to_timeline\n",
    "\n",
    "selected_entropy_lateagg_tl = tensor_to_timeline(selected_entropy_lateagg, model_fps)\n",
    "selected_entropy_lateagg_tl"
   ]
  }
 ],
 "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": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}