faz_experiments.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d7d204a2-1d61-488b-8c5d-25b5cb14e15c",
   "metadata": {},
   "outputs": [],
   "source": [
    "from datetime import datetime\n",
    "import json\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "from search_clustering.pipeline import *\n",
    "from search_clustering.client import *\n",
    "from search_clustering.preprocessing import *\n",
    "from search_clustering.embedding import *\n",
    "from search_clustering.reduction import *\n",
    "from search_clustering.clustering.knn import *\n",
    "from search_clustering.clustering.temporal import *\n",
    "from search_clustering.labeling import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "e4fc5b0d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000\n",
      "200 results\n"
     ]
    }
   ],
   "source": [
    "query = \"energie\"\n",
    "samples = 200\n",
    "use_cached_data = True\n",
    "\n",
    "if not use_cached_data:\n",
    "    es = ElasticClient()\n",
    "\n",
    "    for index in es.client.indices.get_alias():\n",
    "        print(index)\n",
    "\n",
    "    results = es.search(index=\"faz\", field=\"body\", query=query, size=10_000)\n",
    "\n",
    "    with open(f\"datasets/{query.replace(' ', '_')}.json\", \"w\") as json_file:\n",
    "        json.dump(results, json_file)\n",
    "\n",
    "\n",
    "with open(f\"datasets/{query.replace(' ', '_')}.json\", \"r\") as json_file:\n",
    "    results = json.load(json_file)\n",
    "    print(len(results))\n",
    "    results = [\n",
    "        res\n",
    "        for res in results\n",
    "        if res[\"_source\"][\"publication_date\"] is not None\n",
    "        # and \"\\n\" in res[\"_source\"][\"body\"]\n",
    "    ][:samples]\n",
    "\n",
    "\n",
    "print(f\"{len(results)} results\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4db6f6b6",
   "metadata": {},
   "source": [
    "# Paragraph-wise Feature Extraction Performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "fa95e5ba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LDA: 1.911818 s\n",
      "SingleRank: 530.79671 s\n",
      "KeyBERT: 18.769 s\n"
     ]
    }
   ],
   "source": [
    "n = 100\n",
    "docs = ParagraphSplitter().transform(results[:n])\n",
    "\n",
    "\n",
    "def measure_time(extractor: Preprocessing):\n",
    "    start = datetime.now()\n",
    "    extractor.transform(docs)\n",
    "    stop = datetime.now()\n",
    "    delta = stop - start\n",
    "    return delta.total_seconds()\n",
    "\n",
    "\n",
    "# LDA\n",
    "# extractor = ParagraphTopicModeling(language=\"de\")\n",
    "# print(\"LDA:\", measure_time(extractor))\n",
    "print(\"LDA: 1.911818 s\")\n",
    "\n",
    "# SingleRank\n",
    "# extractor = ParagraphGraphKeyphraseExtractor(language=\"de\")\n",
    "# print(\"SingleRank:\", measure_time(extractor)\n",
    "print(\"SingleRank: 530.79671 s\")\n",
    "\n",
    "# KeyBERT\n",
    "# extractor = ParagraphLLMKeyphraseExtractor(language=\"de\")\n",
    "# print(\"KeyBERT:\", measure_time(extractor))\n",
    "print(\"KeyBERT: 18.769 s\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "id": "d79abb7c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(4, 3))\n",
    "\n",
    "x = np.arange(3) / 2\n",
    "heights = [1.91, 530.80, 18.78]\n",
    "labels = [\"LDA\", \"SingleRank\", \"KeyBERT\"]\n",
    "offset = 10\n",
    "\n",
    "plt.bar(x, heights, 0.3)\n",
    "for i in range(3):\n",
    "    plt.text(x[i], heights[i] + offset, heights[i], ha=\"center\")\n",
    "\n",
    "plt.xticks(x, [\"LDA\", \"SingleRank\", \"KeyBERT\"])\n",
    "plt.ylabel(\"Execution Time [s]\")\n",
    "plt.ylim(0, max(heights) + 5 * offset)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "572f3e38",
   "metadata": {},
   "source": [
    "# Clustering Plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "7daab372",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1/6] Preprocessing\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Christopher\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.9_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python39\\site-packages\\sklearn\\feature_extraction\\text.py:409: UserWarning: Your stop_words may be inconsistent with your preprocessing. Tokenizing the stop words generated tokens ['ll', 've'] not in stop_words.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2/6] Embedding\n",
      "[3/6] Reducing Dimensionality\n",
      "[4/6] Clustering\n",
      "[5/6] Labeling\n",
      "[6/6] Visualizing\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "'\\n# KPE 2\\npipe = KNNPipeline(\\n    preprocessing=[\\n        ParagraphSplitter(),\\n        ParagraphLLMKeyphraseExtractor(\"de\"),\\n    ],\\n    embedding=PooledEmbeddings(\"keyphrases\"),\\n    reduction=Umap(),\\n    clustering=clustering,\\n    labeling=FrequentPhrases(language=\"de\", n_phrases=2),\\n)\\n\\ndocs, clusters, labels, score = pipe.fit_transform(results, title=\"KeyBERT (Weighted Pooling)\", legend=False)\\n'"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clustering = HDBSCAN()\n",
    "\n",
    "\"\"\"\n",
    "# LDA 1\n",
    "pipe = KNNPipeline(\n",
    "    preprocessing=[\n",
    "        ParagraphSplitter(),\n",
    "        ParagraphTopicModeling(\"de\"),\n",
    "        ListJoiner(\"topics\"),\n",
    "    ],\n",
    "    embedding=SentenceMiniLM(\"joined\"),\n",
    "    reduction=Umap(),\n",
    "    clustering=clustering,\n",
    "    labeling=FrequentPhrases(language=\"de\"),\n",
    ")\n",
    "docs, clusters, labels, score = pipe.fit_transform(results, title=\"LDA (Concatenation)\", legend=False)\n",
    "\n",
    "# LDA 2\n",
    "pipe = KNNPipeline(\n",
    "    preprocessing=[\n",
    "        ParagraphSplitter(),\n",
    "        ParagraphTopicModeling(\"de\"),\n",
    "    ],\n",
    "    embedding=PooledEmbeddings(\"topics\"),\n",
    "    reduction=Umap(),\n",
    "    clustering=clustering,\n",
    "    labeling=FrequentPhrases(language=\"de\"),\n",
    ")\n",
    "pipe.fit_transform(results, title=\"LDA (Weighted Pooling)\", legend=False)\n",
    "\"\"\"\n",
    "# KPE 1\n",
    "pipe = KNNPipeline(\n",
    "    preprocessing=[\n",
    "        ParagraphSplitter(),\n",
    "        ParagraphLLMKeyphraseExtractor(\"de\"),\n",
    "        ListJoiner(\"keyphrases\"),\n",
    "    ],\n",
    "    embedding=SentenceMiniLM(\"joined\"),\n",
    "    reduction=Umap(),\n",
    "    clustering=clustering,\n",
    "    labeling=FrequentPhrases(language=\"de\", n_phrases=2),\n",
    ")\n",
    "docs, clusters, labels, score = pipe.fit_transform(\n",
    "    results, title=\"KeyBERT (Concatenation)\", legend=False\n",
    ")\n",
    "\"\"\"\n",
    "# KPE 2\n",
    "pipe = KNNPipeline(\n",
    "    preprocessing=[\n",
    "        ParagraphSplitter(),\n",
    "        ParagraphLLMKeyphraseExtractor(\"de\"),\n",
    "    ],\n",
    "    embedding=PooledEmbeddings(\"keyphrases\"),\n",
    "    reduction=Umap(),\n",
    "    clustering=clustering,\n",
    "    labeling=FrequentPhrases(language=\"de\", n_phrases=2),\n",
    ")\n",
    "\n",
    "docs, clusters, labels, score = pipe.fit_transform(results, title=\"KeyBERT (Weighted Pooling)\", legend=False)\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6979e131",
   "metadata": {},
   "source": [
    "# Hierarchical Clustering Plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "f4048970",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[6/6] Visualizing\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[6/6] Visualizing\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[6/6] Visualizing\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[6/6] Visualizing\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "linkages = [\"ward\", \"complete\", \"average\", \"single\"]\n",
    "\n",
    "for linkage in linkages:\n",
    "    pipe = KNNPipeline(\n",
    "        preprocessing=[\n",
    "            ParagraphSplitter(),\n",
    "            ParagraphLLMKeyphraseExtractor(\"de\"),\n",
    "        ],\n",
    "        embedding=PooledEmbeddings(\"keyphrases\"),\n",
    "        reduction=Umap(),\n",
    "        clustering=HierarchicalClustering(linkage=linkage),\n",
    "        labeling=FrequentPhrases(language=\"de\", n_phrases=2),\n",
    "    )\n",
    "    docs, clusters, labels, score = pipe.fit_transform(\n",
    "        results, verbose=False, query=\"heidelberg\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "50dd6b44",
   "metadata": {},
   "source": [
    "# BM 25 Robustness"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "id": "e18ab7dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "25.0"
      ]
     },
     "execution_count": 275,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bm25_docs = [doc[\"_score\"] for doc in docs]\n",
    "bm25_clusters = []\n",
    "for c in sorted(set(clusters)):\n",
    "    cluster = np.take(docs, np.where(clusters == c)[0])\n",
    "    bm25_clusters.append(max([doc[\"_score\"] for doc in cluster]))\n",
    "\n",
    "qs = np.linspace(0, 1)\n",
    "counts = []\n",
    "\n",
    "for q in qs:\n",
    "    indices = (np.array(bm25_clusters) < np.quantile(bm25_docs, q)).nonzero()[0]\n",
    "    counts.append(len(indices))\n",
    "\n",
    "plt.figure(figsize=(5, 4))\n",
    "plt.step(qs[::-1] * len(docs), max(counts) - np.array(counts))\n",
    "plt.vlines(\n",
    "    len(docs) / 4,\n",
    "    0,\n",
    "    max(counts),\n",
    "    color=\"C1\",\n",
    "    linestyle=\"dashed\",\n",
    "    label=f\"BM25 $Q_1$ @ {np.quantile(bm25_docs, 0.25):.2f}\",\n",
    ")\n",
    "plt.vlines(\n",
    "    len(docs) / 2,\n",
    "    0,\n",
    "    max(counts),\n",
    "    color=\"C2\",\n",
    "    linestyle=\"dashed\",\n",
    "    label=f\"BM25 $Q_2$ @ {np.median(bm25_docs):.2f}\",\n",
    ")\n",
    "plt.xlabel(\"Top $n$ documents\")\n",
    "plt.ylabel(\"Number of Topics\")\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()\n",
    "\n",
    "0.25 * len(docs)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "45d5c200",
   "metadata": {},
   "source": [
    "# HDBSCAN n_clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "79954831",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9\n"
     ]
    }
   ],
   "source": [
    "queries = [\"heidelberg\", \"virus\", \"energie\"]\n",
    "width = 0.25\n",
    "x = np.arange(6)\n",
    "colors = [\"C0\", \"C1\", \"C2\"]\n",
    "\n",
    "heights_concat_100 = [18, 17, 15]\n",
    "heights_concat_200 = [35, 33, 29]\n",
    "heights_concat_500 = [30, 17, 7]\n",
    "heights_weighted_100 = [15, 18, 13]\n",
    "heights_weighted_200 = [28, 31, 20]\n",
    "heights_weighted_500 = [32, 3, 22]\n",
    "\n",
    "heights_heidelberg = [18, 35, 30, 15, 28, 32]\n",
    "heights_virus = [17, 33, 17, 18, 31, 3]\n",
    "heights_energie = [15, 29, 7, 13, 20, 22]\n",
    "\n",
    "plt.figure(figsize=(5, 4))\n",
    "\n",
    "plt.bar(x - width, heights_heidelberg, width, color=\"C0\", label=\"heidelberg\")\n",
    "plt.bar(x, heights_virus, width, color=\"C1\", label=\"virus\")\n",
    "plt.bar(x + width, heights_energie, width, color=\"C2\", label=\"energie\")\n",
    "\n",
    "plt.xticks(x, [100, \"200\\nConcatenation\", 500, 100, \"200\\nWeighted Pooling\", 500])\n",
    "plt.ylim(0, 45)\n",
    "plt.ylabel(\"Number of Clusters\")\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.show()\n",
    "\n",
    "print(len(labels))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11396cfb",
   "metadata": {},
   "source": [
    "# Concatenation vs Weighted Pooling Performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 340,
   "id": "db4f9a47",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2023-04-13 06:34:26,008 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:29,621 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:29,622 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:29,623 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:29,624 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:29,625 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:29,626 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:29,627 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:29,628 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:29,629 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:29,630 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:29,631 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:42,317 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:42,318 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:42,318 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:47,669 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:47,670 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:47,671 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:47,672 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:47,673 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:49,586 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:53,361 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:34:53,598 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,792 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,793 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,794 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,794 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,795 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,795 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,796 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,797 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,798 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,798 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,800 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:10,800 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:11,339 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:11,340 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:11,343 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:11,345 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:13,099 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:13,100 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:13,100 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:15,538 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:15,539 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:16,212 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "2023-04-13 06:35:16,213 Warning: An empty Sentence was created! Are there empty strings in your dataset?\n",
      "7.461612351694522\n"
     ]
    }
   ],
   "source": [
    "paragraphs = ParagraphSplitter().transform(results)\n",
    "paragraphs = ParagraphLLMKeyphraseExtractor(\"de\").transform(paragraphs)\n",
    "\n",
    "start = datetime.now()\n",
    "paragraphs = ListJoiner(\"keyphrases\").transform(paragraphs)\n",
    "SentenceMiniLM(\"joined\").transform(paragraphs)\n",
    "stop = datetime.now()\n",
    "delta_concat = stop - start\n",
    "\n",
    "start = datetime.now()\n",
    "PooledEmbeddings(\"keyphrases\").transform(paragraphs)\n",
    "stop = datetime.now()\n",
    "delta_pooled = stop - start\n",
    "\n",
    "print(delta_pooled / delta_concat)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9647597c",
   "metadata": {},
   "source": [
    "# Temporal Scatterplot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "id": "f3da9353",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "17"
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def scatter_temporal(docs, clusters):\n",
    "    plt.figure(figsize=(4, 3))\n",
    "    idx = sorted(set(clusters))\n",
    "    idx = idx[1:13]  # 12 clusters for nice layout, idx[0] = -1 is noise\n",
    "\n",
    "    for j in range(len(idx)):\n",
    "        indices = np.where(clusters == idx[j])[0]\n",
    "        timestamps = [docs[i][\"_source\"][\"publication_date\"] for i in indices]\n",
    "        timestamps = [\n",
    "            datetime.strptime(stamp, \"%Y-%m-%d %H:%M:%S\") for stamp in timestamps\n",
    "        ]\n",
    "        plt.scatter(\n",
    "            timestamps,\n",
    "            [len(labels) - j - 1 for _ in range(len(timestamps))],\n",
    "            label=labels[idx[j]],\n",
    "            alpha=0.5,\n",
    "        )\n",
    "    plt.yticks([])\n",
    "    plt.legend(loc=\"center left\", bbox_to_anchor=(1, 0.5), fontsize=9)\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "scatter_temporal(results, clusters)\n",
    "\n",
    "other = np.take(results, np.where(clusters == 0)[0])\n",
    "for doc in other:\n",
    "    break\n",
    "    print(doc[\"_source\"][\"title\"])\n",
    "\n",
    "len(labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e553bef5",
   "metadata": {},
   "source": [
    "# Temporal Clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "id": "4605cf7d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1/4] Preprocessing\n",
      "[2/4] Clustering\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:73: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  hits = df[df.date >= boundaries[i]][df.date < boundaries[i + 1]]\n",
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:74: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df[\"cluster\"][hits.index] = i\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[3/4] Labeling\n",
      "[4/4] Visualizing\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "d:\\masterthesis\\search_clustering\\clustering\\temporal.py:132: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  clusters[clusters == i] = mapping[i]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pipe = TemporalPipeline(\n",
    "    DummyPreprocessor(), TemporalClustering(target_bins=len(labels)), TemporalLabeling()\n",
    ")\n",
    "docs_temp, clusters_temp, labels_temp = pipe.fit_transform(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "id": "ec21c661",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ARI: 0.02721320337200871\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(3, 12)"
      ]
     },
     "execution_count": 159,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import adjusted_rand_score\n",
    "from sklearn.model_selection import ParameterGrid\n",
    "\n",
    "\n",
    "def jaccard(x: set, y: set):\n",
    "    return len(x.intersection(y)) / len(x.union(y))\n",
    "\n",
    "\n",
    "print(\"ARI:\", adjusted_rand_score(clusters, clusters_temp))\n",
    "\n",
    "ids_topical = []\n",
    "ids_temporal = []\n",
    "similarities = {}\n",
    "\n",
    "for c in sorted(set(clusters)):\n",
    "    docs_c = np.take(docs, (np.where(clusters == c)[0]))\n",
    "    ids_topical.append(set([doc[\"_id\"] for doc in docs_c]))\n",
    "\n",
    "for c in sorted(set(clusters_temp)):\n",
    "    docs_c = np.take(docs_temp, (np.where(clusters_temp == c)[0]))\n",
    "    ids_temporal.append(set([doc[\"_id\"] for doc in docs_c]))\n",
    "\n",
    "grid = ParameterGrid(\n",
    "    [{\"topical\": sorted(set(clusters)), \"temporal\": sorted(set(clusters_temp))}]\n",
    ")\n",
    "for pair in grid:\n",
    "    set_topical = ids_topical[pair[\"topical\"]]\n",
    "    set_temporal = ids_temporal[pair[\"temporal\"]]\n",
    "    sim = jaccard(set_topical, set_temporal)\n",
    "    similarities[(pair[\"topical\"], pair[\"temporal\"])] = sim\n",
    "\n",
    "sort = sorted(similarities.items(), key=lambda item: item[1], reverse=True)\n",
    "\n",
    "# np.take(docs, (np.where(clusters == 1)[0]))\n",
    "# np.take(docs_temp, (np.where(clusters_temp == 0)[0]))\n",
    "print(labels[sort[0][0]], labels[sort[0][1]], sort[0][0])"
   ]
  }
 ],
 "metadata": {
  "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}