notebooks/analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ita    1290\n",
       "nld    1290\n",
       "ara    1290\n",
       "ukr    1290\n",
       "tur    1290\n",
       "vie    1290\n",
       "Name: lang_id, dtype: int64"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "data = pd.read_csv(\"../data/processed/merged/full_main.tsv\", sep=\"\\t\")\n",
    "data.lang_id.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Uncomment to save filtered_data for reusing in modeling.Rmd\n",
    "#outliers = [\n",
    "#    'flores101-main-1003', 'flores101-main-541', 'flores101-main-1001', 'flores101-main-1021', 'flores101-main-1061', \n",
    "#    'flores101-main-1072', 'flores101-main-1074', 'flores101-main-132', 'flores101-main-243', 'flores101-main-284', \n",
    "#    'flores101-main-31', 'flores101-main-32', 'flores101-main-331', 'flores101-main-332', 'flores101-main-403', 'flores101-main-503', 'flores101-main-412'\n",
    "#]\n",
    "#filtered_data = data[~data.item_id.isin(outliers)]\n",
    "#list_ht_prod = filtered_data[filtered_data.task_type == \"ht\"][[\"lang_id\", \"item_id\", \"words_per_minute\"]].to_dict(orient=\"records\")\n",
    "#dic_ht_prod = {(x[\"lang_id\"], x[\"item_id\"]): x[\"words_per_minute\"] for x in list_ht_prod}\n",
    "#filtered_data[\"delta_prod\"] = filtered_data.apply(lambda x: (x[\"words_per_minute\"] - dic_ht_prod[(x[\"lang_id\"], x[\"item_id\"])]) / abs(dic_ht_prod[(x[\"lang_id\"], x[\"item_id\"])]), axis=1)\n",
    "#filtered_data.to_csv(\"../data/processed/filtered_divemt.tsv\", sep=\"\\t\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['lang_id'] = data['lang_id'].map(\n",
    "    {'ara': 'Arabic', 'ita': 'Italian', 'nld': 'Dutch', 'tur': 'Turkish', 'ukr': 'Ukrain.', 'vie': 'Vietnam.'}\n",
    ")\n",
    "data['task_type'] = data['task_type'].map(\n",
    "    {'ht': 'From Scratch (HT)', 'pe1': 'Google Translate (PE1)', 'pe2': \"mBART-50 (PE2)\"}\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Arabic 2\n",
      "Italian 0\n",
      "Dutch 0\n",
      "Turkish 0\n",
      "Ukrainian 0\n",
      "Vietnamese 15\n",
      "Arabic ['flores101-main-1003', 'flores101-main-541']\n",
      "Italian []\n",
      "Dutch []\n",
      "Turkish []\n",
      "Ukrainian []\n",
      "Vietnamese ['flores101-main-1001', 'flores101-main-1021', 'flores101-main-1061', 'flores101-main-1072', 'flores101-main-1074', 'flores101-main-132', 'flores101-main-243', 'flores101-main-284', 'flores101-main-31', 'flores101-main-32', 'flores101-main-331', 'flores101-main-332', 'flores101-main-403', 'flores101-main-503', 'flores101-main-412']\n"
     ]
    }
   ],
   "source": [
    "def print_outliers(data, field, threshold_value, see_ids=False):\n",
    "    print(\"Arabic\", len(data[(data.lang_id == \"Arabic\") & (data.loc[:, field] > threshold_value)]))\n",
    "    print(\"Italian\", len(data[(data.lang_id == \"Italian\") & (data.loc[:, field] > threshold_value)]))\n",
    "    print(\"Dutch\", len(data[(data.lang_id == \"Dutch\") & (data.loc[:, field] > threshold_value)]))\n",
    "    print(\"Turkish\", len(data[(data.lang_id == \"Turkish\") & (data.loc[:, field] > threshold_value)]))\n",
    "    print(\"Ukrainian\", len(data[(data.lang_id == \"Ukrain.\") & (data.loc[:, field] > threshold_value)]))\n",
    "    print(\"Vietnamese\", len(data[(data.lang_id == \"Vietnam.\") & (data.loc[:, field] > threshold_value)]))\n",
    "    if see_ids:\n",
    "        print(\"Arabic\", list(data[(data.lang_id == \"Arabic\") & (data.loc[:, field] > threshold_value)].item_id))\n",
    "        print(\"Italian\", list(data[(data.lang_id == \"Italian\") & (data.loc[:, field] > threshold_value)].item_id))\n",
    "        print(\"Dutch\", list(data[(data.lang_id == \"Dutch\") & (data.loc[:, field] > threshold_value)].item_id))\n",
    "        print(\"Turkish\", list(data[(data.lang_id == \"Turkish\") & (data.loc[:, field] > threshold_value)].item_id))\n",
    "        print(\"Ukrainian\", list(data[(data.lang_id == \"Ukrain.\") & (data.loc[:, field] > threshold_value)].item_id))\n",
    "        print(\"Vietnamese\", list(data[(data.lang_id == \"Vietnam.\") & (data.loc[:, field] > threshold_value)].item_id))\n",
    "\n",
    "print_outliers(data, \"time_m\", 45, True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dutch       1239\n",
       "Turkish     1239\n",
       "Arabic      1239\n",
       "Vietnam.    1239\n",
       "Italian     1239\n",
       "Ukrain.     1239\n",
       "Name: lang_id, dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "outliers = [\n",
    "    'flores101-main-1003', 'flores101-main-541', 'flores101-main-1001', 'flores101-main-1021', 'flores101-main-1061', \n",
    "    'flores101-main-1072', 'flores101-main-1074', 'flores101-main-132', 'flores101-main-243', 'flores101-main-284', \n",
    "    'flores101-main-31', 'flores101-main-32', 'flores101-main-331', 'flores101-main-332', 'flores101-main-403', 'flores101-main-503', 'flores101-main-412'\n",
    "]\n",
    "filtered_data = data[~data.item_id.isin(outliers)]\n",
    "filtered_data.lang_id.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_191773/2626347442.py:4: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\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",
      "  filtered_data[\"delta_prod\"] = filtered_data.apply(lambda x: (x[\"words_per_minute\"] - dic_ht_prod[(x[\"lang_id\"], x[\"item_id\"])]) / abs(dic_ht_prod[(x[\"lang_id\"], x[\"item_id\"])]), axis=1)\n"
     ]
    }
   ],
   "source": [
    "# Compute productivity gains with respect to translation from scratch for every sentence\n",
    "list_ht_prod = filtered_data[filtered_data.task_type == \"From Scratch (HT)\"][[\"lang_id\", \"item_id\", \"words_per_minute\"]].to_dict(orient=\"records\")\n",
    "dic_ht_prod = {(x[\"lang_id\"], x[\"item_id\"]): x[\"words_per_minute\"] for x in list_ht_prod}\n",
    "filtered_data[\"delta_prod\"] = filtered_data.apply(lambda x: (x[\"words_per_minute\"] - dic_ht_prod[(x[\"lang_id\"], x[\"item_id\"])]) / abs(dic_ht_prod[(x[\"lang_id\"], x[\"item_id\"])]), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Regroup by document\n",
    "doc_grouped = filtered_data.groupby([\"doc_id\", \"subject_id\", \"lang_id\", \"task_type\"]).sum()\n",
    "doc_grouped.reset_index(level=[0,1,2,3], inplace=True)\n",
    "doc_grouped.words_per_minute = doc_grouped.src_len_wrd / doc_grouped.time_m\n",
    "doc_grouped.key_per_char = doc_grouped.k_total / doc_grouped.src_len_chr\n",
    "doc_grouped.time_per_char = doc_grouped.time_s / doc_grouped.src_len_chr\n",
    "doc_grouped.time_per_word = doc_grouped.time_s / doc_grouped.src_len_wrd\n",
    "\n",
    "list_ht_prod_doc = filtered_data[filtered_data.task_type == \"From Scratch (HT)\"].groupby([\"lang_id\", \"doc_id\"]).sum().reset_index(level=[0,1])[[\"lang_id\", \"doc_id\", \"src_len_wrd\", \"time_m\"]].to_dict(orient=\"records\")\n",
    "dic_ht_prod_doc = {(x[\"lang_id\"], x[\"doc_id\"]): x[\"src_len_wrd\"] / x[\"time_m\"] for x in list_ht_prod_doc}\n",
    "doc_grouped.delta_prod = doc_grouped.apply(lambda x: (x[\"words_per_minute\"] - dic_ht_prod_doc[(x[\"lang_id\"], x[\"doc_id\"])]) / abs(dic_ht_prod_doc[(x[\"lang_id\"], x[\"doc_id\"])]), axis=1)\n",
    "doc_grouped.bleu = list(filtered_data.groupby([\"doc_id\", \"subject_id\", \"lang_id\", \"task_type\"]).mean()[\"bleu\"])\n",
    "doc_grouped.chrf = list(filtered_data.groupby([\"doc_id\", \"subject_id\", \"lang_id\", \"task_type\"]).mean()[\"chrf\"])\n",
    "doc_grouped.hter = list(filtered_data.groupby([\"doc_id\", \"subject_id\", \"lang_id\", \"task_type\"]).mean()[\"hter\"])\n",
    "doc_grouped.cer = list(filtered_data.groupby([\"doc_id\", \"subject_id\", \"lang_id\", \"task_type\"]).mean()[\"cer\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plotting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from plotnine import (\n",
    "    ggplot, aes, geom_boxplot, labs, facet_grid, xlab, ylab, scale_y_continuous,\n",
    "    ggtitle, scale_fill_brewer, element_blank, element_line, element_text, theme, annotate, scale_fill_manual, facet_wrap\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 27 rows containing non-finite values.\n",
      "findfont: Font family ['Tahoma'] not found. Falling back to DejaVu Sans.\n",
      "findfont: Font family ['Tahoma'] not found. Falling back to DejaVu Sans.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8777006144464)>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot = (\n",
    "    ggplot(\n",
    "        doc_grouped, aes(\"lang_id\", \"words_per_minute\", fill=\"task_type\")\n",
    "    )\n",
    "    + geom_boxplot()\n",
    "    + xlab(\"Target language\")\n",
    "    + ylab(\"Productivity (words/minute)\")\n",
    "    + scale_y_continuous(breaks=np.arange(0, 105, 10),\n",
    "        limits=[0, 100])\n",
    "    + theme(\n",
    "        legend_position=(0.78, 0.75),\n",
    "        legend_title=element_text(size=10, face=\"bold\", colour=\"black\"),\n",
    "        legend_text=element_text(size=9, colour=\"black\"),\n",
    "        legend_title_align=\"center\",\n",
    "        legend_box_spacing=0.4,\n",
    "        legend_key=element_blank(),\n",
    "        axis_line=element_line(size=1, colour=\"black\"),\n",
    "        panel_grid_major=element_line(colour=\"#d3d3d3\"),\n",
    "        panel_grid_minor=element_blank(),\n",
    "        panel_border=element_blank(),\n",
    "        panel_background=element_blank(),\n",
    "        plot_title=element_text(size=15, family=\"Tahoma\", face=\"bold\"),\n",
    "        text=element_text(family=\"Tahoma\", size=9),\n",
    "        axis_text_x=element_text(colour=\"black\", size=9),\n",
    "        axis_text_y=element_text(colour=\"black\", size=9),\n",
    "        figure_size=(10,4),\n",
    "        axis_title=element_text(size=10, face=\"bold\", colour=\"black\"),\n",
    "    )\n",
    "    + scale_fill_manual(values=[\"#66CCFF\", \"#9999CC\", \"#66CC99\"], name=\"Modality\")\n",
    "    + annotate(\"text\", size=9, color = 'black', x = 1, y = 100, label=\"↑ 1\")\n",
    "    + annotate(\"text\", size=9, color = 'black', x = 2, y = 100, label=\"↑ 3\")\n",
    "    + annotate(\"text\", size=9, color = 'black', x = 3, y = 100, label=\"↑ 11\")\n",
    "    + annotate(\"text\", size=9, color = 'black', x = 4, y = 100, label=\"↑ 12\")\n",
    ")\n",
    "plot\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/ggplot.py:820: PlotnineWarning: Filename: productivity_per_document.pdf\n",
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 27 rows containing non-finite values.\n",
      "findfont: Font family ['Tahoma'] not found. Falling back to DejaVu Sans.\n",
      "findfont: Font family ['Tahoma'] not found. Falling back to DejaVu Sans.\n"
     ]
    }
   ],
   "source": [
    "from plotnine import save_as_pdf_pages\n",
    "\n",
    "save_as_pdf_pages([plot], \"../outputs/productivity_per_document.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 2 rows containing non-finite values.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8777003722993)>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot = (\n",
    "    ggplot(\n",
    "        doc_grouped, aes(\"lang_id\", \"key_per_char\", fill=\"task_type\")\n",
    "    )\n",
    "    + geom_boxplot()\n",
    "    + xlab(\"Target language\")\n",
    "    + ylab(\"Technical effort (keys/char)\")\n",
    "    + scale_y_continuous(breaks=np.arange(0, 6, 1), \n",
    "                     limits=[0, 6.2])\n",
    "    + theme(\n",
    "        legend_position=(0.38, 0.81),\n",
    "        legend_direction=\"horizontal\",\n",
    "        legend_title=element_text(size=10, face=\"bold\", colour=\"black\"),\n",
    "        legend_text=element_text(size=9, colour=\"black\"),\n",
    "        legend_title_align=\"center\",\n",
    "        legend_box_spacing=0.4,\n",
    "        legend_key=element_blank(),\n",
    "        axis_line=element_line(size=1, colour=\"black\"),\n",
    "        panel_grid_major=element_line(colour=\"#d3d3d3\"),\n",
    "        panel_grid_minor=element_blank(),\n",
    "        panel_border=element_blank(),\n",
    "        panel_background=element_blank(),\n",
    "        plot_title=element_text(size=15, family=\"Tahoma\", face=\"bold\"),\n",
    "        text=element_text(family=\"Tahoma\", size=9),\n",
    "        axis_text_x=element_text(colour=\"black\", size=9),\n",
    "        axis_text_y=element_text(colour=\"black\", size=9),\n",
    "        axis_title=element_text(size=10, face=\"bold\", colour=\"black\"),\n",
    "        figure_size=(10,4)\n",
    "    )\n",
    "    + scale_fill_manual(values=[\"#66CCFF\", \"#9999CC\", \"#66CC99\"], name=\"Modality\")\n",
    ")\n",
    "plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/ggplot.py:820: PlotnineWarning: Filename: technical_effort.pdf\n",
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 2 rows containing non-finite values.\n"
     ]
    }
   ],
   "source": [
    "from plotnine import save_as_pdf_pages\n",
    "\n",
    "save_as_pdf_pages([plot], \"../outputs/technical_effort.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 24 rows containing non-finite values.\n",
      "findfont: Font family ['Tahoma'] not found. Falling back to DejaVu Sans.\n",
      "findfont: Font family ['Tahoma'] not found. Falling back to DejaVu Sans.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8777003557548)>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from plotnine import (\n",
    "    ggplot, aes, geom_boxplot, labs, facet_grid, xlab, ylab, scale_y_continuous,\n",
    "    ggtitle, scale_fill_brewer, element_blank, element_line, element_text, theme, annotate, scale_fill_manual\n",
    ")\n",
    "\n",
    "plot = (\n",
    "    ggplot(# time_per_word\n",
    "        doc_grouped, aes(\"lang_id\", \"time_per_word\", fill=\"task_type\")\n",
    "    )\n",
    "    + geom_boxplot()\n",
    "    + xlab(\"Target language\")\n",
    "    + ylab(\"Time per source word (s)\")\n",
    "    + scale_y_continuous(breaks=np.arange(0, 22, 3),\n",
    "        limits=[0, 22])\n",
    "    + theme(\n",
    "        legend_position=(0.5, 0.96),\n",
    "        legend_title=element_blank(),#element_text(size=10, face=\"bold\", colour=\"black\"),\n",
    "        legend_text=element_text(size=10, colour=\"black\"),\n",
    "        legend_title_align=\"center\",\n",
    "        legend_direction=\"horizontal\",\n",
    "        legend_box_spacing=0.05,\n",
    "        legend_key=element_blank(),\n",
    "        axis_line=element_line(size=1, colour=\"black\"),\n",
    "        panel_grid_major=element_line(colour=\"#d3d3d3\"),\n",
    "        panel_grid_minor=element_blank(),\n",
    "        panel_border=element_blank(),\n",
    "        panel_background=element_blank(),\n",
    "        plot_title=element_text(size=15, family=\"Tahoma\", face=\"bold\"),\n",
    "        text=element_text(family=\"Tahoma\", size=10),\n",
    "        axis_text_x=element_text(colour=\"black\", size=10),\n",
    "        axis_title_x = element_blank(),\n",
    "        axis_text_y=element_text(colour=\"black\", size=10),\n",
    "        figure_size=(6,3),\n",
    "        axis_title_y=element_text(size=11, face=\"bold\", colour=\"black\"),\n",
    "        legend_box_margin=0.15,\n",
    "    )\n",
    "    + scale_fill_manual(values=[\"#66CCFF\", \"#9999CC\", \"#66CC99\"], name=\"Modality\")\n",
    "    + annotate(\"text\", size=9, color = 'black', x = 1, y = 22, label=\"↑ 8\")\n",
    "    + annotate(\"text\", size=9, color = 'black', x = 3, y = 22, label=\"↑ 3\")\n",
    "    + annotate(\"text\", size=9, color = 'black', x = 5, y = 22, label=\"↑ 3\")\n",
    "    + annotate(\"text\", size=9, color = 'black', x = 6, y = 22, label=\"↑ 10\")\n",
    ")\n",
    "plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Arabic 8\n",
      "Italian 0\n",
      "Dutch 3\n",
      "Turkish 0\n",
      "Ukrainian 3\n",
      "Vietnamese 10\n"
     ]
    }
   ],
   "source": [
    "print_outliers(doc_grouped, \"time_per_word\", 22)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/ggplot.py:820: PlotnineWarning: Filename: time_per_src_word_small.pdf\n",
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 24 rows containing non-finite values.\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/series.py:1001\u001b[0m, in \u001b[0;36mSeries.__setitem__\u001b[0;34m(self, key, value)\u001b[0m\n\u001b[1;32m   1000\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m-> 1001\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_set_with_engine(key, value)\n\u001b[1;32m   1002\u001b[0m \u001b[39mexcept\u001b[39;00m (\u001b[39mKeyError\u001b[39;00m, \u001b[39mValueError\u001b[39;00m):\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/series.py:1034\u001b[0m, in \u001b[0;36mSeries._set_with_engine\u001b[0;34m(self, key, value)\u001b[0m\n\u001b[1;32m   1032\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_set_with_engine\u001b[39m(\u001b[39mself\u001b[39m, key, value):\n\u001b[1;32m   1033\u001b[0m     \u001b[39m# fails with AttributeError for IntervalIndex\u001b[39;00m\n\u001b[0;32m-> 1034\u001b[0m     loc \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mindex\u001b[39m.\u001b[39;49m_engine\u001b[39m.\u001b[39;49mget_loc(key)\n\u001b[1;32m   1035\u001b[0m     validate_numeric_casting(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdtype, value)\n",
      "File \u001b[0;32mpandas/_libs/index.pyx:70\u001b[0m, in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
      "File \u001b[0;32mpandas/_libs/index.pyx:75\u001b[0m, in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;31mTypeError\u001b[0m: '0    False\n1    False\n2    False\nName: x, dtype: bool' is an invalid key",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m/home/gsarti/projects/divemt/notebooks/analysis.ipynb Cell 16'\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      <a href='vscode-notebook-cell:/home/gsarti/projects/divemt/notebooks/analysis.ipynb#ch0000014?line=0'>1</a>\u001b[0m \u001b[39mfrom\u001b[39;00m \u001b[39mplotnine\u001b[39;00m \u001b[39mimport\u001b[39;00m save_as_pdf_pages\n\u001b[0;32m----> <a href='vscode-notebook-cell:/home/gsarti/projects/divemt/notebooks/analysis.ipynb#ch0000014?line=2'>3</a>\u001b[0m save_as_pdf_pages([plot], \u001b[39m\"\u001b[39;49m\u001b[39mtime_per_src_word_small.pdf\u001b[39;49m\u001b[39m\"\u001b[39;49m)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/ggplot.py:825\u001b[0m, in \u001b[0;36msave_as_pdf_pages\u001b[0;34m(plots, filename, path, verbose, **kwargs)\u001b[0m\n\u001b[1;32m    822\u001b[0m \u001b[39mwith\u001b[39;00m PdfPages(filename) \u001b[39mas\u001b[39;00m pdf:\n\u001b[1;32m    823\u001b[0m     \u001b[39m# Re-add the first element to the iterator, if it was removed\u001b[39;00m\n\u001b[1;32m    824\u001b[0m     \u001b[39mfor\u001b[39;00m plot \u001b[39min\u001b[39;00m plots:\n\u001b[0;32m--> 825\u001b[0m         fig \u001b[39m=\u001b[39m plot\u001b[39m.\u001b[39;49mdraw()\n\u001b[1;32m    826\u001b[0m         \u001b[39m# Save as a page in the PDF file\u001b[39;00m\n\u001b[1;32m    827\u001b[0m         pdf\u001b[39m.\u001b[39msavefig(fig, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mfig_kwargs)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/ggplot.py:214\u001b[0m, in \u001b[0;36mggplot.draw\u001b[0;34m(self, return_ggplot, show)\u001b[0m\n\u001b[1;32m    211\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_resize_panels()\n\u001b[1;32m    213\u001b[0m \u001b[39m# Drawing\u001b[39;00m\n\u001b[0;32m--> 214\u001b[0m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_draw_layers()\n\u001b[1;32m    215\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_draw_labels()\n\u001b[1;32m    216\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_draw_breaks_and_labels()\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/ggplot.py:391\u001b[0m, in \u001b[0;36mggplot._draw_layers\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    387\u001b[0m \u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m    388\u001b[0m \u001b[39mDraw the main plot(s) onto the axes.\u001b[39;00m\n\u001b[1;32m    389\u001b[0m \u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m    390\u001b[0m \u001b[39m# Draw the geoms\u001b[39;00m\n\u001b[0;32m--> 391\u001b[0m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mlayers\u001b[39m.\u001b[39;49mdraw(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mlayout, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mcoordinates)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:65\u001b[0m, in \u001b[0;36mLayers.draw\u001b[0;34m(self, layout, coord)\u001b[0m\n\u001b[1;32m     63\u001b[0m \u001b[39mfor\u001b[39;00m i, l \u001b[39min\u001b[39;00m \u001b[39menumerate\u001b[39m(\u001b[39mself\u001b[39m, start\u001b[39m=\u001b[39m\u001b[39m1\u001b[39m):\n\u001b[1;32m     64\u001b[0m     l\u001b[39m.\u001b[39mzorder \u001b[39m=\u001b[39m i\n\u001b[0;32m---> 65\u001b[0m     l\u001b[39m.\u001b[39;49mdraw(layout, coord)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:404\u001b[0m, in \u001b[0;36mlayer.draw\u001b[0;34m(self, layout, coord)\u001b[0m\n\u001b[1;32m    401\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdata \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mgeom\u001b[39m.\u001b[39mhandle_na(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdata)\n\u001b[1;32m    402\u001b[0m \u001b[39m# At this point each layer must have the data\u001b[39;00m\n\u001b[1;32m    403\u001b[0m \u001b[39m# that is created by the plot build process\u001b[39;00m\n\u001b[0;32m--> 404\u001b[0m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mgeom\u001b[39m.\u001b[39;49mdraw_layer(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mdata, layout, coord, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mparams)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/geoms/geom.py:237\u001b[0m, in \u001b[0;36mgeom.draw_layer\u001b[0;34m(self, data, layout, coord, **params)\u001b[0m\n\u001b[1;32m    235\u001b[0m panel_params \u001b[39m=\u001b[39m layout\u001b[39m.\u001b[39mpanel_params[ploc]\n\u001b[1;32m    236\u001b[0m ax \u001b[39m=\u001b[39m layout\u001b[39m.\u001b[39maxs[ploc]\n\u001b[0;32m--> 237\u001b[0m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mdraw_panel(pdata, panel_params, coord, ax, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mparams)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/geoms/geom.py:272\u001b[0m, in \u001b[0;36mgeom.draw_panel\u001b[0;34m(self, data, panel_params, coord, ax, **params)\u001b[0m\n\u001b[1;32m    270\u001b[0m \u001b[39mfor\u001b[39;00m _, gdata \u001b[39min\u001b[39;00m data\u001b[39m.\u001b[39mgroupby(\u001b[39m'\u001b[39m\u001b[39mgroup\u001b[39m\u001b[39m'\u001b[39m):\n\u001b[1;32m    271\u001b[0m     gdata\u001b[39m.\u001b[39mreset_index(inplace\u001b[39m=\u001b[39m\u001b[39mTrue\u001b[39;00m, drop\u001b[39m=\u001b[39m\u001b[39mTrue\u001b[39;00m)\n\u001b[0;32m--> 272\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mdraw_group(gdata, panel_params, coord, ax, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mparams)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/geoms/geom_boxplot.py:174\u001b[0m, in \u001b[0;36mgeom_boxplot.draw_group\u001b[0;34m(data, panel_params, coord, ax, **params)\u001b[0m\n\u001b[1;32m    172\u001b[0m     outliers[\u001b[39m'\u001b[39m\u001b[39msize\u001b[39m\u001b[39m'\u001b[39m] \u001b[39m=\u001b[39m outlier_value(\u001b[39m'\u001b[39m\u001b[39msize\u001b[39m\u001b[39m'\u001b[39m)\n\u001b[1;32m    173\u001b[0m     outliers[\u001b[39m'\u001b[39m\u001b[39mstroke\u001b[39m\u001b[39m'\u001b[39m] \u001b[39m=\u001b[39m outlier_value(\u001b[39m'\u001b[39m\u001b[39mstroke\u001b[39m\u001b[39m'\u001b[39m)\n\u001b[0;32m--> 174\u001b[0m     geom_point\u001b[39m.\u001b[39;49mdraw_group(outliers, panel_params,\n\u001b[1;32m    175\u001b[0m                           coord, ax, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mparams)\n\u001b[1;32m    177\u001b[0m \u001b[39m# plot\u001b[39;00m\n\u001b[1;32m    178\u001b[0m geom_segment\u001b[39m.\u001b[39mdraw_group(whiskers, panel_params,\n\u001b[1;32m    179\u001b[0m                         coord, ax, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mparams)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/geoms/geom_point.py:35\u001b[0m, in \u001b[0;36mgeom_point.draw_group\u001b[0;34m(data, panel_params, coord, ax, **params)\u001b[0m\n\u001b[1;32m     33\u001b[0m \u001b[39m@staticmethod\u001b[39m\n\u001b[1;32m     34\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mdraw_group\u001b[39m(data, panel_params, coord, ax, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mparams):\n\u001b[0;32m---> 35\u001b[0m     data \u001b[39m=\u001b[39m coord\u001b[39m.\u001b[39;49mtransform(data, panel_params)\n\u001b[1;32m     36\u001b[0m     units \u001b[39m=\u001b[39m \u001b[39m'\u001b[39m\u001b[39mshape\u001b[39m\u001b[39m'\u001b[39m\n\u001b[1;32m     37\u001b[0m     \u001b[39mfor\u001b[39;00m _, udata \u001b[39min\u001b[39;00m data\u001b[39m.\u001b[39mgroupby(units, dropna\u001b[39m=\u001b[39m\u001b[39mFalse\u001b[39;00m):\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/coords/coord_cartesian.py:40\u001b[0m, in \u001b[0;36mcoord_cartesian.transform\u001b[0;34m(self, data, panel_params, munch)\u001b[0m\n\u001b[1;32m     37\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39msquish_infinite_y\u001b[39m(data):\n\u001b[1;32m     38\u001b[0m     \u001b[39mreturn\u001b[39;00m squish_infinite(data, \u001b[39mrange\u001b[39m\u001b[39m=\u001b[39mpanel_params\u001b[39m.\u001b[39my\u001b[39m.\u001b[39mrange)\n\u001b[0;32m---> 40\u001b[0m \u001b[39mreturn\u001b[39;00m transform_position(data, squish_infinite_x, squish_infinite_y)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/positions/position.py:97\u001b[0m, in \u001b[0;36mposition.transform_position\u001b[0;34m(data, trans_x, trans_y)\u001b[0m\n\u001b[1;32m     95\u001b[0m \u001b[39mif\u001b[39;00m trans_x:\n\u001b[1;32m     96\u001b[0m     xs \u001b[39m=\u001b[39m [name \u001b[39mfor\u001b[39;00m name \u001b[39min\u001b[39;00m data\u001b[39m.\u001b[39mcolumns \u001b[39mif\u001b[39;00m name \u001b[39min\u001b[39;00m X]\n\u001b[0;32m---> 97\u001b[0m     data[xs] \u001b[39m=\u001b[39m data[xs]\u001b[39m.\u001b[39;49mapply(trans_x)\n\u001b[1;32m     99\u001b[0m \u001b[39mif\u001b[39;00m trans_y:\n\u001b[1;32m    100\u001b[0m     ys \u001b[39m=\u001b[39m [name \u001b[39mfor\u001b[39;00m name \u001b[39min\u001b[39;00m data\u001b[39m.\u001b[39mcolumns \u001b[39mif\u001b[39;00m name \u001b[39min\u001b[39;00m Y]\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/frame.py:7552\u001b[0m, in \u001b[0;36mDataFrame.apply\u001b[0;34m(self, func, axis, raw, result_type, args, **kwds)\u001b[0m\n\u001b[1;32m   7541\u001b[0m \u001b[39mfrom\u001b[39;00m \u001b[39mpandas\u001b[39;00m\u001b[39m.\u001b[39;00m\u001b[39mcore\u001b[39;00m\u001b[39m.\u001b[39;00m\u001b[39mapply\u001b[39;00m \u001b[39mimport\u001b[39;00m frame_apply\n\u001b[1;32m   7543\u001b[0m op \u001b[39m=\u001b[39m frame_apply(\n\u001b[1;32m   7544\u001b[0m     \u001b[39mself\u001b[39m,\n\u001b[1;32m   7545\u001b[0m     func\u001b[39m=\u001b[39mfunc,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   7550\u001b[0m     kwds\u001b[39m=\u001b[39mkwds,\n\u001b[1;32m   7551\u001b[0m )\n\u001b[0;32m-> 7552\u001b[0m \u001b[39mreturn\u001b[39;00m op\u001b[39m.\u001b[39;49mget_result()\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/apply.py:185\u001b[0m, in \u001b[0;36mFrameApply.get_result\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    182\u001b[0m \u001b[39melif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mraw:\n\u001b[1;32m    183\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mapply_raw()\n\u001b[0;32m--> 185\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mapply_standard()\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/apply.py:276\u001b[0m, in \u001b[0;36mFrameApply.apply_standard\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    275\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mapply_standard\u001b[39m(\u001b[39mself\u001b[39m):\n\u001b[0;32m--> 276\u001b[0m     results, res_index \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mapply_series_generator()\n\u001b[1;32m    278\u001b[0m     \u001b[39m# wrap results\u001b[39;00m\n\u001b[1;32m    279\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mwrap_results(results, res_index)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/apply.py:305\u001b[0m, in \u001b[0;36mFrameApply.apply_series_generator\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    302\u001b[0m \u001b[39mwith\u001b[39;00m option_context(\u001b[39m\"\u001b[39m\u001b[39mmode.chained_assignment\u001b[39m\u001b[39m\"\u001b[39m, \u001b[39mNone\u001b[39;00m):\n\u001b[1;32m    303\u001b[0m     \u001b[39mfor\u001b[39;00m i, v \u001b[39min\u001b[39;00m \u001b[39menumerate\u001b[39m(series_gen):\n\u001b[1;32m    304\u001b[0m         \u001b[39m# ignore SettingWithCopy here in case the user mutates\u001b[39;00m\n\u001b[0;32m--> 305\u001b[0m         results[i] \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mf(v)\n\u001b[1;32m    306\u001b[0m         \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(results[i], ABCSeries):\n\u001b[1;32m    307\u001b[0m             \u001b[39m# If we have a view on v, we need to make a copy because\u001b[39;00m\n\u001b[1;32m    308\u001b[0m             \u001b[39m#  series_generator will swap out the underlying data\u001b[39;00m\n\u001b[1;32m    309\u001b[0m             results[i] \u001b[39m=\u001b[39m results[i]\u001b[39m.\u001b[39mcopy(deep\u001b[39m=\u001b[39m\u001b[39mFalse\u001b[39;00m)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/plotnine/coords/coord_cartesian.py:35\u001b[0m, in \u001b[0;36mcoord_cartesian.transform.<locals>.squish_infinite_x\u001b[0;34m(data)\u001b[0m\n\u001b[1;32m     34\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39msquish_infinite_x\u001b[39m(data):\n\u001b[0;32m---> 35\u001b[0m     \u001b[39mreturn\u001b[39;00m squish_infinite(data, \u001b[39mrange\u001b[39;49m\u001b[39m=\u001b[39;49mpanel_params\u001b[39m.\u001b[39;49mx\u001b[39m.\u001b[39;49mrange)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/mizani/bounds.py:233\u001b[0m, in \u001b[0;36msquish_infinite\u001b[0;34m(x, range)\u001b[0m\n\u001b[1;32m    230\u001b[0m     x \u001b[39m=\u001b[39m np\u001b[39m.\u001b[39masarray(x)\n\u001b[1;32m    232\u001b[0m x[x \u001b[39m==\u001b[39m \u001b[39m-\u001b[39mnp\u001b[39m.\u001b[39minf] \u001b[39m=\u001b[39m \u001b[39mrange\u001b[39m[\u001b[39m0\u001b[39m]\n\u001b[0;32m--> 233\u001b[0m x[x \u001b[39m==\u001b[39m np\u001b[39m.\u001b[39minf] \u001b[39m=\u001b[39m \u001b[39mrange\u001b[39m[\u001b[39m1\u001b[39m]\n\u001b[1;32m    235\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39misinstance\u001b[39m(x, xtype):\n\u001b[1;32m    236\u001b[0m     x \u001b[39m=\u001b[39m xtype(x)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/series.py:1021\u001b[0m, in \u001b[0;36mSeries.__setitem__\u001b[0;34m(self, key, value)\u001b[0m\n\u001b[1;32m   1019\u001b[0m key \u001b[39m=\u001b[39m np\u001b[39m.\u001b[39masarray(key, dtype\u001b[39m=\u001b[39m\u001b[39mbool\u001b[39m)\n\u001b[1;32m   1020\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m-> 1021\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_where(\u001b[39m~\u001b[39;49mkey, value, inplace\u001b[39m=\u001b[39;49m\u001b[39mTrue\u001b[39;49;00m)\n\u001b[1;32m   1022\u001b[0m \u001b[39mexcept\u001b[39;00m InvalidIndexError:\n\u001b[1;32m   1023\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39miloc[key] \u001b[39m=\u001b[39m value\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/generic.py:8760\u001b[0m, in \u001b[0;36mNDFrame._where\u001b[0;34m(self, cond, other, inplace, axis, level, errors, try_cast)\u001b[0m\n\u001b[1;32m   8758\u001b[0m \u001b[39m# make sure we are boolean\u001b[39;00m\n\u001b[1;32m   8759\u001b[0m fill_value \u001b[39m=\u001b[39m \u001b[39mbool\u001b[39m(inplace)\n\u001b[0;32m-> 8760\u001b[0m cond \u001b[39m=\u001b[39m cond\u001b[39m.\u001b[39;49mfillna(fill_value)\n\u001b[1;32m   8762\u001b[0m msg \u001b[39m=\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mBoolean array expected for the condition, not \u001b[39m\u001b[39m{dtype}\u001b[39;00m\u001b[39m\"\u001b[39m\n\u001b[1;32m   8764\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m cond\u001b[39m.\u001b[39mempty:\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/series.py:4530\u001b[0m, in \u001b[0;36mSeries.fillna\u001b[0;34m(self, value, method, axis, inplace, limit, downcast)\u001b[0m\n\u001b[1;32m   4520\u001b[0m \u001b[39m@doc\u001b[39m(NDFrame\u001b[39m.\u001b[39mfillna, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39m_shared_doc_kwargs)\n\u001b[1;32m   4521\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mfillna\u001b[39m(\n\u001b[1;32m   4522\u001b[0m     \u001b[39mself\u001b[39m,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   4528\u001b[0m     downcast\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m,\n\u001b[1;32m   4529\u001b[0m ) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m Optional[\u001b[39m\"\u001b[39m\u001b[39mSeries\u001b[39m\u001b[39m\"\u001b[39m]:\n\u001b[0;32m-> 4530\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39msuper\u001b[39;49m()\u001b[39m.\u001b[39;49mfillna(\n\u001b[1;32m   4531\u001b[0m         value\u001b[39m=\u001b[39;49mvalue,\n\u001b[1;32m   4532\u001b[0m         method\u001b[39m=\u001b[39;49mmethod,\n\u001b[1;32m   4533\u001b[0m         axis\u001b[39m=\u001b[39;49maxis,\n\u001b[1;32m   4534\u001b[0m         inplace\u001b[39m=\u001b[39;49minplace,\n\u001b[1;32m   4535\u001b[0m         limit\u001b[39m=\u001b[39;49mlimit,\n\u001b[1;32m   4536\u001b[0m         downcast\u001b[39m=\u001b[39;49mdowncast,\n\u001b[1;32m   4537\u001b[0m     )\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/generic.py:6061\u001b[0m, in \u001b[0;36mNDFrame.fillna\u001b[0;34m(self, value, method, axis, inplace, limit, downcast)\u001b[0m\n\u001b[1;32m   6054\u001b[0m     \u001b[39melse\u001b[39;00m:\n\u001b[1;32m   6055\u001b[0m         \u001b[39mraise\u001b[39;00m \u001b[39mTypeError\u001b[39;00m(\n\u001b[1;32m   6056\u001b[0m             \u001b[39m'\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mvalue\u001b[39m\u001b[39m\"\u001b[39m\u001b[39m parameter must be a scalar, dict \u001b[39m\u001b[39m'\u001b[39m\n\u001b[1;32m   6057\u001b[0m             \u001b[39m\"\u001b[39m\u001b[39mor Series, but you passed a \u001b[39m\u001b[39m\"\u001b[39m\n\u001b[1;32m   6058\u001b[0m             \u001b[39mf\u001b[39m\u001b[39m'\u001b[39m\u001b[39m\"\u001b[39m\u001b[39m{\u001b[39;00m\u001b[39mtype\u001b[39m(value)\u001b[39m.\u001b[39m\u001b[39m__name__\u001b[39m\u001b[39m}\u001b[39;00m\u001b[39m\"\u001b[39m\u001b[39m'\u001b[39m\n\u001b[1;32m   6059\u001b[0m         )\n\u001b[0;32m-> 6061\u001b[0m     new_data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_mgr\u001b[39m.\u001b[39;49mfillna(\n\u001b[1;32m   6062\u001b[0m         value\u001b[39m=\u001b[39;49mvalue, limit\u001b[39m=\u001b[39;49mlimit, inplace\u001b[39m=\u001b[39;49minplace, downcast\u001b[39m=\u001b[39;49mdowncast\n\u001b[1;32m   6063\u001b[0m     )\n\u001b[1;32m   6065\u001b[0m \u001b[39melif\u001b[39;00m \u001b[39misinstance\u001b[39m(value, (\u001b[39mdict\u001b[39m, ABCSeries)):\n\u001b[1;32m   6066\u001b[0m     \u001b[39mif\u001b[39;00m axis \u001b[39m==\u001b[39m \u001b[39m1\u001b[39m:\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/internals/managers.py:594\u001b[0m, in \u001b[0;36mBlockManager.fillna\u001b[0;34m(self, value, limit, inplace, downcast)\u001b[0m\n\u001b[1;32m    593\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mfillna\u001b[39m(\u001b[39mself\u001b[39m, value, limit, inplace: \u001b[39mbool\u001b[39m, downcast) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mBlockManager\u001b[39m\u001b[39m\"\u001b[39m:\n\u001b[0;32m--> 594\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mapply(\n\u001b[1;32m    595\u001b[0m         \u001b[39m\"\u001b[39;49m\u001b[39mfillna\u001b[39;49m\u001b[39m\"\u001b[39;49m, value\u001b[39m=\u001b[39;49mvalue, limit\u001b[39m=\u001b[39;49mlimit, inplace\u001b[39m=\u001b[39;49minplace, downcast\u001b[39m=\u001b[39;49mdowncast\n\u001b[1;32m    596\u001b[0m     )\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/internals/managers.py:409\u001b[0m, in \u001b[0;36mBlockManager.apply\u001b[0;34m(self, f, align_keys, **kwargs)\u001b[0m\n\u001b[1;32m    407\u001b[0m         applied \u001b[39m=\u001b[39m b\u001b[39m.\u001b[39mapply(f, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\n\u001b[1;32m    408\u001b[0m     \u001b[39melse\u001b[39;00m:\n\u001b[0;32m--> 409\u001b[0m         applied \u001b[39m=\u001b[39m \u001b[39mgetattr\u001b[39;49m(b, f)(\u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m    410\u001b[0m     result_blocks \u001b[39m=\u001b[39m _extend_blocks(applied, result_blocks)\n\u001b[1;32m    412\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mlen\u001b[39m(result_blocks) \u001b[39m==\u001b[39m \u001b[39m0\u001b[39m:\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/internals/blocks.py:376\u001b[0m, in \u001b[0;36mBlock.fillna\u001b[0;34m(self, value, limit, inplace, downcast)\u001b[0m\n\u001b[1;32m    370\u001b[0m \u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m    371\u001b[0m \u001b[39mfillna on the block with the value. If we fail, then convert to\u001b[39;00m\n\u001b[1;32m    372\u001b[0m \u001b[39mObjectBlock and try again\u001b[39;00m\n\u001b[1;32m    373\u001b[0m \u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m    374\u001b[0m inplace \u001b[39m=\u001b[39m validate_bool_kwarg(inplace, \u001b[39m\"\u001b[39m\u001b[39minplace\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[0;32m--> 376\u001b[0m mask \u001b[39m=\u001b[39m isna(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mvalues)\n\u001b[1;32m    377\u001b[0m \u001b[39mif\u001b[39;00m limit \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m    378\u001b[0m     limit \u001b[39m=\u001b[39m libalgos\u001b[39m.\u001b[39m_validate_limit(\u001b[39mNone\u001b[39;00m, limit\u001b[39m=\u001b[39mlimit)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/dtypes/missing.py:124\u001b[0m, in \u001b[0;36misna\u001b[0;34m(obj)\u001b[0m\n\u001b[1;32m     47\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39misna\u001b[39m(obj):\n\u001b[1;32m     48\u001b[0m     \u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m     49\u001b[0m \u001b[39m    Detect missing values for an array-like object.\u001b[39;00m\n\u001b[1;32m     50\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    122\u001b[0m \u001b[39m    Name: 1, dtype: bool\u001b[39;00m\n\u001b[1;32m    123\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m--> 124\u001b[0m     \u001b[39mreturn\u001b[39;00m _isna(obj)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/dtypes/missing.py:157\u001b[0m, in \u001b[0;36m_isna\u001b[0;34m(obj, inf_as_na)\u001b[0m\n\u001b[1;32m    155\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mFalse\u001b[39;00m\n\u001b[1;32m    156\u001b[0m \u001b[39melif\u001b[39;00m \u001b[39misinstance\u001b[39m(obj, (ABCSeries, np\u001b[39m.\u001b[39mndarray, ABCIndexClass, ABCExtensionArray)):\n\u001b[0;32m--> 157\u001b[0m     \u001b[39mreturn\u001b[39;00m _isna_ndarraylike(obj, inf_as_na\u001b[39m=\u001b[39;49minf_as_na)\n\u001b[1;32m    158\u001b[0m \u001b[39melif\u001b[39;00m \u001b[39misinstance\u001b[39m(obj, ABCDataFrame):\n\u001b[1;32m    159\u001b[0m     \u001b[39mreturn\u001b[39;00m obj\u001b[39m.\u001b[39misna()\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/dtypes/missing.py:219\u001b[0m, in \u001b[0;36m_isna_ndarraylike\u001b[0;34m(obj, inf_as_na)\u001b[0m\n\u001b[1;32m    217\u001b[0m \u001b[39melif\u001b[39;00m is_string_dtype(dtype):\n\u001b[1;32m    218\u001b[0m     result \u001b[39m=\u001b[39m _isna_string_dtype(values, dtype, inf_as_na\u001b[39m=\u001b[39minf_as_na)\n\u001b[0;32m--> 219\u001b[0m \u001b[39melif\u001b[39;00m needs_i8_conversion(dtype):\n\u001b[1;32m    220\u001b[0m     \u001b[39m# this is the NaT pattern\u001b[39;00m\n\u001b[1;32m    221\u001b[0m     result \u001b[39m=\u001b[39m values\u001b[39m.\u001b[39mview(\u001b[39m\"\u001b[39m\u001b[39mi8\u001b[39m\u001b[39m\"\u001b[39m) \u001b[39m==\u001b[39m iNaT\n\u001b[1;32m    222\u001b[0m \u001b[39melse\u001b[39;00m:\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/dtypes/common.py:1218\u001b[0m, in \u001b[0;36mneeds_i8_conversion\u001b[0;34m(arr_or_dtype)\u001b[0m\n\u001b[1;32m   1214\u001b[0m \u001b[39mif\u001b[39;00m arr_or_dtype \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m   1215\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mFalse\u001b[39;00m\n\u001b[1;32m   1216\u001b[0m \u001b[39mreturn\u001b[39;00m (\n\u001b[1;32m   1217\u001b[0m     is_datetime_or_timedelta_dtype(arr_or_dtype)\n\u001b[0;32m-> 1218\u001b[0m     \u001b[39mor\u001b[39;00m is_datetime64tz_dtype(arr_or_dtype)\n\u001b[1;32m   1219\u001b[0m     \u001b[39mor\u001b[39;00m is_period_dtype(arr_or_dtype)\n\u001b[1;32m   1220\u001b[0m )\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/dtypes/common.py:419\u001b[0m, in \u001b[0;36mis_datetime64tz_dtype\u001b[0;34m(arr_or_dtype)\u001b[0m\n\u001b[1;32m    417\u001b[0m \u001b[39mif\u001b[39;00m arr_or_dtype \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m    418\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mFalse\u001b[39;00m\n\u001b[0;32m--> 419\u001b[0m \u001b[39mreturn\u001b[39;00m DatetimeTZDtype\u001b[39m.\u001b[39;49mis_dtype(arr_or_dtype)\n",
      "File \u001b[0;32m~/projects/venv/lib/python3.8/site-packages/pandas/core/dtypes/base.py:280\u001b[0m, in \u001b[0;36mExtensionDtype.is_dtype\u001b[0;34m(cls, dtype)\u001b[0m\n\u001b[1;32m    256\u001b[0m \u001b[39m@classmethod\u001b[39m\n\u001b[1;32m    257\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mis_dtype\u001b[39m(\u001b[39mcls\u001b[39m, dtype: \u001b[39mobject\u001b[39m) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m \u001b[39mbool\u001b[39m:\n\u001b[1;32m    258\u001b[0m     \u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m    259\u001b[0m \u001b[39m    Check if we match 'dtype'.\u001b[39;00m\n\u001b[1;32m    260\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    278\u001b[0m \u001b[39m       conditions is true for ``dtype.dtype``.\u001b[39;00m\n\u001b[1;32m    279\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m--> 280\u001b[0m     dtype \u001b[39m=\u001b[39m \u001b[39mgetattr\u001b[39;49m(dtype, \u001b[39m\"\u001b[39;49m\u001b[39mdtype\u001b[39;49m\u001b[39m\"\u001b[39;49m, dtype)\n\u001b[1;32m    282\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(dtype, (ABCSeries, ABCIndexClass, ABCDataFrame, np\u001b[39m.\u001b[39mdtype)):\n\u001b[1;32m    283\u001b[0m         \u001b[39m# https://github.com/pandas-dev/pandas/issues/22960\u001b[39;00m\n\u001b[1;32m    284\u001b[0m         \u001b[39m# avoid passing data to `construct_from_string`. This could\u001b[39;00m\n\u001b[1;32m    285\u001b[0m         \u001b[39m# cause a FutureWarning from numpy about failing elementwise\u001b[39;00m\n\u001b[1;32m    286\u001b[0m         \u001b[39m# comparison from, e.g., comparing DataFrame == 'category'.\u001b[39;00m\n\u001b[1;32m    287\u001b[0m         \u001b[39mreturn\u001b[39;00m \u001b[39mFalse\u001b[39;00m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "from plotnine import save_as_pdf_pages\n",
    "\n",
    "save_as_pdf_pages([plot], \"../outputs/time_per_src_word_small.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_groups = filtered_data[[\"subject_id\", \"task_type\"]]\n",
    "full_groups = full_groups[full_groups[\"task_type\"] != \"ht\"]\n",
    "filter_groups = filtered_data[filtered_data.tot_edits < 1][[\"subject_id\", \"task_type\"]]\n",
    "filter_groups = filter_groups[filter_groups[\"task_type\"] != \"ht\"]\n",
    "x = filter_groups.value_counts() / full_groups.value_counts()\n",
    "x.fillna(0).to_csv(\"../outputs/out_edits_eval.csv\")\n",
    "x = pd.read_csv(\"../outputs/out_edits_eval.csv\")\n",
    "x[\"lang\"] = x.subject_id.str.split(\"_\").map(lambda x: x[0])\n",
    "x['lang'] = x['lang'].map(\n",
    "    {'ara': 'Arabic', 'ita': 'Italian', 'nld': 'Dutch', 'tur': 'Turkish', 'ukr': 'Ukrain.', 'vie': 'Vietnam.'}\n",
    ")\n",
    "x[\"subject_id\"] = x.subject_id.str.split(\"_\").map(lambda x: x[1])\n",
    "x.rename({\"0\": \"score\"}, axis=1, inplace=True)\n",
    "x.loc[x[(x.lang == \"Italian\") & (x.subject_id == \"t5\")].index, \"subject_id\"] = \"t2\"\n",
    "x.loc[x[(x.lang == \"Italian\") & (x.subject_id == \"t4\")].index, \"subject_id\"] = \"t3\"\n",
    "x[\"score\"] = x[\"score\"] * 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8736203067350)>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from plotnine import geom_col, facet_grid\n",
    "\n",
    "y = x.copy()\n",
    "y['task_type'] = y['task_type'].map(\n",
    "    {'From Scratch (HT)': 'ht', 'Google Translate (PE1)': 'Google (PE1)', \"mBART-50 (PE2)\": 'mBART-50 (PE2)'}\n",
    ")\n",
    "\n",
    "plot = (\n",
    "    ggplot(\n",
    "        y[y.task_type != \"ht\"], aes(\"subject_id\", \"score\")\n",
    "    )\n",
    "    + geom_col()\n",
    "    + facet_grid(\"task_type ~ lang\")\n",
    "    + xlab(\"Subject x Language\")\n",
    "    + ylab(\"% Errorless MT Outputs\")\n",
    "    + theme(\n",
    "        axis_title=element_text(size=10, face=\"bold\", colour=\"black\"),\n",
    "        axis_text_x=element_blank(),\n",
    "        axis_text_y=element_text(size=9, colour=\"black\"),\n",
    "        axis_ticks=element_blank(),\n",
    "        figure_size=(6,4)\n",
    "    )\n",
    "    + scale_fill_brewer(type=\"qual\", palette=\"Accent\", name=\"Translation setting\")\n",
    ")\n",
    "plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/ggplot.py:820: PlotnineWarning: Filename: ../outputs/errorless_outputs_small.pdf\n"
     ]
    }
   ],
   "source": [
    "from plotnine import save_as_pdf_pages\n",
    "\n",
    "save_as_pdf_pages([plot], \"../outputs/errorless_outputs_small.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 5 rows containing non-finite values.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8787520642033)>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from plotnine import guide_legend, guides\n",
    "\n",
    "plot = (\n",
    "    ggplot(\n",
    "        filtered_data[filtered_data.task_type.isin([\"Google Translate (PE1)\", \"mBART-50 (PE2)\"])], aes(\"task_type\", \"hter\", fill=\"lang_id\")\n",
    "    )\n",
    "    + geom_boxplot()\n",
    "    + xlab(\"NMT System\")\n",
    "    + ylab(\"HTER (edits/words)\")\n",
    "    + scale_y_continuous(limits=[0, 150], breaks=np.arange(0, 155, 25))\n",
    "    + theme(\n",
    "        legend_position=(0.50, 0.90),\n",
    "        legend_direction=\"horizontal\",\n",
    "        legend_title_align=\"center\",\n",
    "        legend_box_spacing=0.05,\n",
    "        legend_key=element_blank(),\n",
    "        axis_line=element_line(size=1, colour=\"black\"),\n",
    "        panel_grid_major=element_line(colour=\"#d3d3d3\"),\n",
    "        panel_grid_minor=element_blank(),\n",
    "        panel_border=element_blank(),\n",
    "        panel_background=element_blank(),\n",
    "        plot_title=element_text(size=15, family=\"Tahoma\", face=\"bold\"),\n",
    "        #text=element_text(family=\"Tahoma\", size=13),\n",
    "        axis_text_x=element_text(colour=\"black\", size=9),\n",
    "        axis_text_y=element_text(colour=\"black\", size=9),\n",
    "        figure_size=(6,4),\n",
    "        axis_title_y=element_text(size=10, face=\"bold\", colour=\"black\"),\n",
    "        axis_title_x=element_blank(),\n",
    "        legend_title=element_blank(),\n",
    "        legend_box_margin=0.15,\n",
    "    )\n",
    "    + scale_fill_brewer(type=\"qual\", palette=\"Accent\", name=\"Language\")\n",
    "    + guides(fill = guide_legend(nrow=1))\n",
    ")\n",
    "plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/ggplot.py:820: PlotnineWarning: Filename: ../outputs/hter_per_system.pdf\n",
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 5 rows containing non-finite values.\n"
     ]
    }
   ],
   "source": [
    "from plotnine import save_as_pdf_pages\n",
    "\n",
    "save_as_pdf_pages([plot], \"../outputs/hter_per_system.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8736204743687)>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from plotnine import guide_legend, guides\n",
    "\n",
    "plot = (\n",
    "    ggplot(\n",
    "        filtered_data[filtered_data.task_type.isin([\"Google Translate (PE1)\", \"mBART-50 (PE2)\"])], aes(\"task_type\", \"cer\", fill=\"lang_id\")\n",
    "    )\n",
    "    + geom_boxplot()\n",
    "    + xlab(\"NMT System\")\n",
    "    + ylab(\"CharacTER\")\n",
    "    + scale_y_continuous(limits=[0, 1.1], breaks=np.arange(0, 1.1, 0.1))\n",
    "    + theme(\n",
    "        legend_position=(0.50, 0.85),\n",
    "        legend_direction=\"horizontal\",\n",
    "        legend_title_align=\"center\",\n",
    "        legend_box_spacing=0.05,\n",
    "        legend_key=element_blank(),\n",
    "        axis_line=element_line(size=1, colour=\"black\"),\n",
    "        panel_grid_major=element_line(colour=\"#d3d3d3\"),\n",
    "        panel_grid_minor=element_blank(),\n",
    "        panel_border=element_blank(),\n",
    "        panel_background=element_blank(),\n",
    "        plot_title=element_text(size=15, family=\"Tahoma\", face=\"bold\"),\n",
    "        #text=element_text(family=\"Tahoma\", size=13),\n",
    "        axis_text_x=element_text(colour=\"black\", size=9),\n",
    "        axis_text_y=element_text(colour=\"black\", size=9),\n",
    "        figure_size=(6,4),\n",
    "        axis_title_y=element_text(size=10, face=\"bold\", colour=\"black\"),\n",
    "        axis_title_x=element_blank(),\n",
    "        legend_title=element_blank(),\n",
    "        legend_box_margin=0.15,\n",
    "    )\n",
    "    + scale_fill_brewer(type=\"qual\", palette=\"Accent\", name=\"Language\")\n",
    "    + guides(fill = guide_legend(nrow=1))\n",
    ")\n",
    "plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/ggplot.py:820: PlotnineWarning: Filename: ../outputs/cer_per_system.pdf\n"
     ]
    }
   ],
   "source": [
    "from plotnine import save_as_pdf_pages\n",
    "\n",
    "save_as_pdf_pages([plot], \"../outputs/cer_per_system.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "doc_grouped[[\"lang_id\", \"subject_single_id\"]] = doc_grouped.subject_id.str.split(\"_\", expand=True)\n",
    "doc_grouped['lang_id'] = doc_grouped['lang_id'].map({'ara': 'Arabic', 'ita': 'Italian', 'nld': 'Dutch', 'tur': 'Turkish', 'ukr': 'Ukrainian', 'vie': 'Vietnamese'})\n",
    "doc_grouped.subject_single_id.replace({'t4': 't3', 't5': 't2'}, inplace=True)\n",
    "doc_grouped.subject_single_id.replace({'t1': 'Subject 1', 't2': 'Subject 2', 't3': 'Subject 3'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 24 rows containing non-finite values.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1200 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8728312338755)>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot = (\n",
    "    ggplot(\n",
    "        doc_grouped, aes(\"subject_single_id\", \"time_per_word\", fill=\"task_type\")\n",
    "    )\n",
    "    + geom_boxplot()\n",
    "    + scale_fill_manual(values=[\"#66CCFF\", \"#9999CC\", \"#66CC99\"], name=\"Modality\")\n",
    "    + facet_wrap(\"~ lang_id\", nrow=3)\n",
    "    + ylab(\"Time per source word (s)\")\n",
    "    + xlab(\"Subject ID\")\n",
    "    + scale_y_continuous(breaks=np.arange(0, 22, 3), \n",
    "                     limits=[0, 22])\n",
    "    + theme(\n",
    "        legend_position=(0.50, 0.93),\n",
    "        legend_direction=\"horizontal\",\n",
    "        legend_title_align=\"center\",\n",
    "        legend_box_spacing=0.4,\n",
    "        legend_title=element_text(size=10, face=\"bold\", colour=\"black\"),\n",
    "        legend_key=element_blank(),\n",
    "        text=element_text(family=\"Tahoma\", size=9),\n",
    "        axis_text_y=element_text(colour=\"black\", size=9),\n",
    "        figure_size=(10,12),\n",
    "        axis_title=element_text(size=10, face=\"bold\", colour=\"black\"),\n",
    "    )\n",
    ")\n",
    "plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/ggplot.py:820: PlotnineWarning: Filename: ../outputs/trans_time_per_word_per_mode.pdf\n",
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 24 rows containing non-finite values.\n"
     ]
    }
   ],
   "source": [
    "from plotnine import save_as_pdf_pages\n",
    "\n",
    "save_as_pdf_pages([plot], \"../outputs/trans_time_per_word_per_mode.pdf\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Scores and Correlations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>lang_id</th>\n",
       "      <th>task_type</th>\n",
       "      <th>words_per_minute</th>\n",
       "      <th>delta_prod</th>\n",
       "      <th>hter</th>\n",
       "      <th>cer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Arabic</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>13.125376</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Arabic</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>21.726673</td>\n",
       "      <td>0.841105</td>\n",
       "      <td>19.485400</td>\n",
       "      <td>0.096628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Arabic</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>16.287586</td>\n",
       "      <td>0.097658</td>\n",
       "      <td>41.387667</td>\n",
       "      <td>0.270816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Dutch</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>13.621550</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Dutch</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>28.704447</td>\n",
       "      <td>1.185477</td>\n",
       "      <td>14.285667</td>\n",
       "      <td>0.102075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Dutch</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>21.708249</td>\n",
       "      <td>0.606589</td>\n",
       "      <td>32.422750</td>\n",
       "      <td>0.238839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Italian</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>8.760813</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Italian</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>18.600289</td>\n",
       "      <td>0.962913</td>\n",
       "      <td>19.192000</td>\n",
       "      <td>0.150653</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Italian</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>15.619779</td>\n",
       "      <td>0.949084</td>\n",
       "      <td>31.052750</td>\n",
       "      <td>0.233580</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Turkish</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>17.944462</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Turkish</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>25.545050</td>\n",
       "      <td>0.339056</td>\n",
       "      <td>19.888800</td>\n",
       "      <td>0.131753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Turkish</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>20.998652</td>\n",
       "      <td>0.124148</td>\n",
       "      <td>43.438000</td>\n",
       "      <td>0.296503</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Ukrain.</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>7.994399</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Ukrain.</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>12.293725</td>\n",
       "      <td>0.713842</td>\n",
       "      <td>51.159333</td>\n",
       "      <td>0.358460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Ukrain.</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>9.817031</td>\n",
       "      <td>0.139370</td>\n",
       "      <td>68.205000</td>\n",
       "      <td>0.498664</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Vietnam.</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>10.172442</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Vietnam.</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>13.047575</td>\n",
       "      <td>0.317728</td>\n",
       "      <td>34.497500</td>\n",
       "      <td>0.289745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Vietnam.</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>11.092330</td>\n",
       "      <td>0.231328</td>\n",
       "      <td>43.421000</td>\n",
       "      <td>0.359985</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     lang_id               task_type  words_per_minute  delta_prod       hter  \\\n",
       "0     Arabic       From Scratch (HT)         13.125376    0.000000        NaN   \n",
       "1     Arabic  Google Translate (PE1)         21.726673    0.841105  19.485400   \n",
       "2     Arabic          mBART-50 (PE2)         16.287586    0.097658  41.387667   \n",
       "3      Dutch       From Scratch (HT)         13.621550    0.000000        NaN   \n",
       "4      Dutch  Google Translate (PE1)         28.704447    1.185477  14.285667   \n",
       "5      Dutch          mBART-50 (PE2)         21.708249    0.606589  32.422750   \n",
       "6    Italian       From Scratch (HT)          8.760813    0.000000        NaN   \n",
       "7    Italian  Google Translate (PE1)         18.600289    0.962913  19.192000   \n",
       "8    Italian          mBART-50 (PE2)         15.619779    0.949084  31.052750   \n",
       "9    Turkish       From Scratch (HT)         17.944462    0.000000        NaN   \n",
       "10   Turkish  Google Translate (PE1)         25.545050    0.339056  19.888800   \n",
       "11   Turkish          mBART-50 (PE2)         20.998652    0.124148  43.438000   \n",
       "12   Ukrain.       From Scratch (HT)          7.994399    0.000000        NaN   \n",
       "13   Ukrain.  Google Translate (PE1)         12.293725    0.713842  51.159333   \n",
       "14   Ukrain.          mBART-50 (PE2)          9.817031    0.139370  68.205000   \n",
       "15  Vietnam.       From Scratch (HT)         10.172442    0.000000        NaN   \n",
       "16  Vietnam.  Google Translate (PE1)         13.047575    0.317728  34.497500   \n",
       "17  Vietnam.          mBART-50 (PE2)         11.092330    0.231328  43.421000   \n",
       "\n",
       "         cer  \n",
       "0        NaN  \n",
       "1   0.096628  \n",
       "2   0.270816  \n",
       "3        NaN  \n",
       "4   0.102075  \n",
       "5   0.238839  \n",
       "6        NaN  \n",
       "7   0.150653  \n",
       "8   0.233580  \n",
       "9        NaN  \n",
       "10  0.131753  \n",
       "11  0.296503  \n",
       "12       NaN  \n",
       "13  0.358460  \n",
       "14  0.498664  \n",
       "15       NaN  \n",
       "16  0.289745  \n",
       "17  0.359985  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lang_grouped_median = doc_grouped.groupby([\"lang_id\", \"task_type\"]).median()\n",
    "lang_grouped_median.reset_index(level=[0,1], inplace=True)\n",
    "lang_grouped_median[[\"lang_id\", \"task_type\", \"words_per_minute\", \"delta_prod\", \"hter\", \"cer\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>lang_id</th>\n",
       "      <th>task_type</th>\n",
       "      <th>words_per_minute</th>\n",
       "      <th>delta_prod</th>\n",
       "      <th>hter</th>\n",
       "      <th>cer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Arabic</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>15.140944</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Arabic</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>26.685869</td>\n",
       "      <td>1.490138</td>\n",
       "      <td>22.781066</td>\n",
       "      <td>0.126949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Arabic</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>16.952308</td>\n",
       "      <td>0.463584</td>\n",
       "      <td>41.996365</td>\n",
       "      <td>0.284544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Dutch</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>15.332657</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Dutch</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>44.861575</td>\n",
       "      <td>3.808754</td>\n",
       "      <td>20.776179</td>\n",
       "      <td>0.159465</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Dutch</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>27.256527</td>\n",
       "      <td>1.325363</td>\n",
       "      <td>34.399926</td>\n",
       "      <td>0.271256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Italian</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>9.946033</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Italian</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>26.552826</td>\n",
       "      <td>2.614879</td>\n",
       "      <td>22.472132</td>\n",
       "      <td>0.168796</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Italian</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>19.867604</td>\n",
       "      <td>1.513266</td>\n",
       "      <td>32.817708</td>\n",
       "      <td>0.241942</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Turkish</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>19.324286</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Turkish</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>64.665544</td>\n",
       "      <td>3.167948</td>\n",
       "      <td>21.671344</td>\n",
       "      <td>0.151297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Turkish</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>25.963002</td>\n",
       "      <td>0.600537</td>\n",
       "      <td>44.567938</td>\n",
       "      <td>0.313614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Ukrain.</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>8.896697</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Ukrain.</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>14.164355</td>\n",
       "      <td>0.977267</td>\n",
       "      <td>51.394015</td>\n",
       "      <td>0.363699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Ukrain.</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>11.477867</td>\n",
       "      <td>0.561070</td>\n",
       "      <td>66.528662</td>\n",
       "      <td>0.480691</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Vietnam.</td>\n",
       "      <td>From Scratch (HT)</td>\n",
       "      <td>10.490802</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Vietnam.</td>\n",
       "      <td>Google Translate (PE1)</td>\n",
       "      <td>14.867669</td>\n",
       "      <td>0.975514</td>\n",
       "      <td>35.136314</td>\n",
       "      <td>0.290238</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Vietnam.</td>\n",
       "      <td>mBART-50 (PE2)</td>\n",
       "      <td>12.563109</td>\n",
       "      <td>0.538952</td>\n",
       "      <td>43.926409</td>\n",
       "      <td>0.360651</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     lang_id               task_type  words_per_minute  delta_prod       hter  \\\n",
       "0     Arabic       From Scratch (HT)         15.140944    0.000000        NaN   \n",
       "1     Arabic  Google Translate (PE1)         26.685869    1.490138  22.781066   \n",
       "2     Arabic          mBART-50 (PE2)         16.952308    0.463584  41.996365   \n",
       "3      Dutch       From Scratch (HT)         15.332657    0.000000        NaN   \n",
       "4      Dutch  Google Translate (PE1)         44.861575    3.808754  20.776179   \n",
       "5      Dutch          mBART-50 (PE2)         27.256527    1.325363  34.399926   \n",
       "6    Italian       From Scratch (HT)          9.946033    0.000000        NaN   \n",
       "7    Italian  Google Translate (PE1)         26.552826    2.614879  22.472132   \n",
       "8    Italian          mBART-50 (PE2)         19.867604    1.513266  32.817708   \n",
       "9    Turkish       From Scratch (HT)         19.324286    0.000000        NaN   \n",
       "10   Turkish  Google Translate (PE1)         64.665544    3.167948  21.671344   \n",
       "11   Turkish          mBART-50 (PE2)         25.963002    0.600537  44.567938   \n",
       "12   Ukrain.       From Scratch (HT)          8.896697    0.000000        NaN   \n",
       "13   Ukrain.  Google Translate (PE1)         14.164355    0.977267  51.394015   \n",
       "14   Ukrain.          mBART-50 (PE2)         11.477867    0.561070  66.528662   \n",
       "15  Vietnam.       From Scratch (HT)         10.490802    0.000000        NaN   \n",
       "16  Vietnam.  Google Translate (PE1)         14.867669    0.975514  35.136314   \n",
       "17  Vietnam.          mBART-50 (PE2)         12.563109    0.538952  43.926409   \n",
       "\n",
       "         cer  \n",
       "0        NaN  \n",
       "1   0.126949  \n",
       "2   0.284544  \n",
       "3        NaN  \n",
       "4   0.159465  \n",
       "5   0.271256  \n",
       "6        NaN  \n",
       "7   0.168796  \n",
       "8   0.241942  \n",
       "9        NaN  \n",
       "10  0.151297  \n",
       "11  0.313614  \n",
       "12       NaN  \n",
       "13  0.363699  \n",
       "14  0.480691  \n",
       "15       NaN  \n",
       "16  0.290238  \n",
       "17  0.360651  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lang_grouped_mean = doc_grouped.groupby([\"lang_id\", \"task_type\"]).mean()\n",
    "lang_grouped_mean.reset_index(level=[0,1], inplace=True)\n",
    "lang_grouped_mean[[\"lang_id\", \"task_type\", \"words_per_minute\", \"delta_prod\", \"hter\", \"cer\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>lang_id</th>\n",
       "      <th>hter</th>\n",
       "      <th>cer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Arabic</td>\n",
       "      <td>32.388716</td>\n",
       "      <td>0.205746</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Dutch</td>\n",
       "      <td>27.588052</td>\n",
       "      <td>0.215361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Italian</td>\n",
       "      <td>27.644920</td>\n",
       "      <td>0.205369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Turkish</td>\n",
       "      <td>33.119641</td>\n",
       "      <td>0.232456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Ukrain.</td>\n",
       "      <td>58.961338</td>\n",
       "      <td>0.422195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Vietnam.</td>\n",
       "      <td>39.531362</td>\n",
       "      <td>0.325444</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    lang_id       hter       cer\n",
       "0    Arabic  32.388716  0.205746\n",
       "1     Dutch  27.588052  0.215361\n",
       "2   Italian  27.644920  0.205369\n",
       "3   Turkish  33.119641  0.232456\n",
       "4   Ukrain.  58.961338  0.422195\n",
       "5  Vietnam.  39.531362  0.325444"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lang_grouped_only_lang_mean = doc_grouped.groupby([\"lang_id\"]).mean()\n",
    "lang_grouped_only_lang_mean.reset_index(level=[0], inplace=True)\n",
    "lang_grouped_only_lang_mean[[\"lang_id\", \"hter\", \"cer\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>bleu</th>\n",
       "      <th>chrf</th>\n",
       "      <th>ter</th>\n",
       "      <th>comet</th>\n",
       "      <th>prod</th>\n",
       "      <th>delta_prod</th>\n",
       "      <th>hter</th>\n",
       "      <th>cer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>bleu</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.254851</td>\n",
       "      <td>-0.992012</td>\n",
       "      <td>0.067340</td>\n",
       "      <td>-0.298409</td>\n",
       "      <td>-0.517914</td>\n",
       "      <td>0.127038</td>\n",
       "      <td>0.254436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chrf</th>\n",
       "      <td>0.254851</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.328335</td>\n",
       "      <td>0.601047</td>\n",
       "      <td>0.449949</td>\n",
       "      <td>0.016493</td>\n",
       "      <td>-0.491775</td>\n",
       "      <td>-0.591422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ter</th>\n",
       "      <td>-0.992012</td>\n",
       "      <td>-0.328335</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.189357</td>\n",
       "      <td>0.219336</td>\n",
       "      <td>0.481177</td>\n",
       "      <td>-0.092242</td>\n",
       "      <td>-0.213708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comet</th>\n",
       "      <td>0.067340</td>\n",
       "      <td>0.601047</td>\n",
       "      <td>-0.189357</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.649162</td>\n",
       "      <td>0.226509</td>\n",
       "      <td>-0.179088</td>\n",
       "      <td>-0.211636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>prod</th>\n",
       "      <td>-0.298409</td>\n",
       "      <td>0.449949</td>\n",
       "      <td>0.219336</td>\n",
       "      <td>0.649162</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.825964</td>\n",
       "      <td>-0.657686</td>\n",
       "      <td>-0.654199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>delta_prod</th>\n",
       "      <td>-0.517914</td>\n",
       "      <td>0.016493</td>\n",
       "      <td>0.481177</td>\n",
       "      <td>0.226509</td>\n",
       "      <td>0.825964</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.734373</td>\n",
       "      <td>-0.677249</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hter</th>\n",
       "      <td>0.127038</td>\n",
       "      <td>-0.491775</td>\n",
       "      <td>-0.092242</td>\n",
       "      <td>-0.179088</td>\n",
       "      <td>-0.657686</td>\n",
       "      <td>-0.734373</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.967526</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cer</th>\n",
       "      <td>0.254436</td>\n",
       "      <td>-0.591422</td>\n",
       "      <td>-0.213708</td>\n",
       "      <td>-0.211636</td>\n",
       "      <td>-0.654199</td>\n",
       "      <td>-0.677249</td>\n",
       "      <td>0.967526</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                bleu      chrf       ter     comet      prod  delta_prod  \\\n",
       "bleu        1.000000  0.254851 -0.992012  0.067340 -0.298409   -0.517914   \n",
       "chrf        0.254851  1.000000 -0.328335  0.601047  0.449949    0.016493   \n",
       "ter        -0.992012 -0.328335  1.000000 -0.189357  0.219336    0.481177   \n",
       "comet       0.067340  0.601047 -0.189357  1.000000  0.649162    0.226509   \n",
       "prod       -0.298409  0.449949  0.219336  0.649162  1.000000    0.825964   \n",
       "delta_prod -0.517914  0.016493  0.481177  0.226509  0.825964    1.000000   \n",
       "hter        0.127038 -0.491775 -0.092242 -0.179088 -0.657686   -0.734373   \n",
       "cer         0.254436 -0.591422 -0.213708 -0.211636 -0.654199   -0.677249   \n",
       "\n",
       "                hter       cer  \n",
       "bleu        0.127038  0.254436  \n",
       "chrf       -0.491775 -0.591422  \n",
       "ter        -0.092242 -0.213708  \n",
       "comet      -0.179088 -0.211636  \n",
       "prod       -0.657686 -0.654199  \n",
       "delta_prod -0.734373 -0.677249  \n",
       "hter        1.000000  0.967526  \n",
       "cer         0.967526  1.000000  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_google = pd.DataFrame({\n",
    "    \"bleu\": [34.1, 29.1, 32.8, 35, 31.1, 45.1],\n",
    "    \"chrf\": [65.6, 60, 61.4, 65.5, 59.8, 61.9],\n",
    "    \"ter\": [52.8, 58.5, 53.6, 50.4, 55.9, 41.8],\n",
    "    \"comet\": [0.737, 0.667, 0.781, 1.000, 0.758, 0.724],\n",
    "    \"prod\": [26.685, 44.86, 26.55, 64.66, 14.16, 14.86],\n",
    "    \"delta_prod\": [1.49, 3.80, 2.61, 3.16, 0.97, 0.98],\n",
    "    \"hter\": [22.78, 20.78, 22.47, 21.67, 51.39, 35.14],\n",
    "    \"cer\": [0.127, 0.159, 0.169, 0.151, 0.364, 0.290]\n",
    "})\n",
    "df_google.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>bleu</th>\n",
       "      <th>chrf</th>\n",
       "      <th>ter</th>\n",
       "      <th>comet</th>\n",
       "      <th>prod</th>\n",
       "      <th>delta_prod</th>\n",
       "      <th>hter</th>\n",
       "      <th>cer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>bleu</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.612593</td>\n",
       "      <td>-0.974188</td>\n",
       "      <td>0.138430</td>\n",
       "      <td>-0.417451</td>\n",
       "      <td>0.050335</td>\n",
       "      <td>-0.074185</td>\n",
       "      <td>0.164183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chrf</th>\n",
       "      <td>0.612593</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.656037</td>\n",
       "      <td>0.544794</td>\n",
       "      <td>0.342025</td>\n",
       "      <td>0.668030</td>\n",
       "      <td>-0.491327</td>\n",
       "      <td>-0.318828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ter</th>\n",
       "      <td>-0.974188</td>\n",
       "      <td>-0.656037</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.140016</td>\n",
       "      <td>0.286952</td>\n",
       "      <td>-0.113750</td>\n",
       "      <td>0.277238</td>\n",
       "      <td>0.046728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comet</th>\n",
       "      <td>0.138430</td>\n",
       "      <td>0.544794</td>\n",
       "      <td>-0.140016</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.266192</td>\n",
       "      <td>0.079479</td>\n",
       "      <td>0.033419</td>\n",
       "      <td>0.048996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>prod</th>\n",
       "      <td>-0.417451</td>\n",
       "      <td>0.342025</td>\n",
       "      <td>0.286952</td>\n",
       "      <td>0.266192</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.512679</td>\n",
       "      <td>-0.624904</td>\n",
       "      <td>-0.671635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>delta_prod</th>\n",
       "      <td>0.050335</td>\n",
       "      <td>0.668030</td>\n",
       "      <td>-0.113750</td>\n",
       "      <td>0.079479</td>\n",
       "      <td>0.512679</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.640778</td>\n",
       "      <td>-0.598382</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hter</th>\n",
       "      <td>-0.074185</td>\n",
       "      <td>-0.491327</td>\n",
       "      <td>0.277238</td>\n",
       "      <td>0.033419</td>\n",
       "      <td>-0.624904</td>\n",
       "      <td>-0.640778</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.966830</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cer</th>\n",
       "      <td>0.164183</td>\n",
       "      <td>-0.318828</td>\n",
       "      <td>0.046728</td>\n",
       "      <td>0.048996</td>\n",
       "      <td>-0.671635</td>\n",
       "      <td>-0.598382</td>\n",
       "      <td>0.966830</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                bleu      chrf       ter     comet      prod  delta_prod  \\\n",
       "bleu        1.000000  0.612593 -0.974188  0.138430 -0.417451    0.050335   \n",
       "chrf        0.612593  1.000000 -0.656037  0.544794  0.342025    0.668030   \n",
       "ter        -0.974188 -0.656037  1.000000 -0.140016  0.286952   -0.113750   \n",
       "comet       0.138430  0.544794 -0.140016  1.000000  0.266192    0.079479   \n",
       "prod       -0.417451  0.342025  0.286952  0.266192  1.000000    0.512679   \n",
       "delta_prod  0.050335  0.668030 -0.113750  0.079479  0.512679    1.000000   \n",
       "hter       -0.074185 -0.491327  0.277238  0.033419 -0.624904   -0.640778   \n",
       "cer         0.164183 -0.318828  0.046728  0.048996 -0.671635   -0.598382   \n",
       "\n",
       "                hter       cer  \n",
       "bleu       -0.074185  0.164183  \n",
       "chrf       -0.491327 -0.318828  \n",
       "ter         0.277238  0.046728  \n",
       "comet       0.033419  0.048996  \n",
       "prod       -0.624904 -0.671635  \n",
       "delta_prod -0.640778 -0.598382  \n",
       "hter        1.000000  0.966830  \n",
       "cer         0.966830  1.000000  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_mbart = pd.DataFrame({\n",
    "    \"bleu\": [17, 22.6, 24.4, 18.8, 21.9, 34.7],\n",
    "    \"chrf\": [48.5, 53.9, 54.7, 52.7, 50.7, 54],\n",
    "    \"ter\": [69.1, 63.7, 61.2, 67.5, 67.9, 50.7],\n",
    "    \"comet\": [0.452, 0.532, 0.648, 0.755, 0.587, 0.608],\n",
    "    \"prod\": [16.95, 27.26, 19.87, 25.96, 11.47, 12.56],\n",
    "    \"delta_prod\": [0.46, 1.33, 1.51, 0.60, 0.56, 0.53],\n",
    "    \"hter\": [42, 34.40, 32.82, 44.57, 66.53, 43.93],\n",
    "    \"cer\": [0.285, 0.271, 0.242, 0.314, 0.481, 0.361]\n",
    "    \n",
    "})\n",
    "df_mbart.corr()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Questionnaires"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "pre = pd.read_csv(\"../data/questionnaire/pre_task.tsv\", sep=\"\\t\")\n",
    "post = pd.read_csv(\"../data/questionnaire/post_task.tsv\", sep=\"\\t\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmIAAAEFCAYAAABAT4uZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA2QUlEQVR4nO3deVhUZfsH8O/MICOyDPtmiqIm5C6kZS645ZLmlgtqCvL6szRDzVxwrzBfU3JJkzBBy6VMSzPX3PN1QzO1EFMjS0H2VQNlzu8PLyaHRQaY4WFmvp/r4tKzPec+N4eHm+ecOUcmSZIEIiIiIqp2ctEBEBEREZkrFmJEREREgrAQIyIiIhKEhRgRERGRICzEiIiIiARhIUZEREQkCAsxIiIiIkFYiBEREREJwkKMiIiISBAWYkRERESCsBAjIiIiEoSFGBEREZEgLMSIiIiIBGEhRkRERCQICzEjkpubi8jISOTm5ooOpUZifkrHvJSNuakY5qvimDMqDwsxI5Kbm4uoqCj+QJeB+Skd81I25qZimK+KY86oPCzEiIiIiASRSZIkiQ6C/uXv76+XdmJjY/XSji4aNGiAdevWoXfv3iWW3b59G8899xzS0tKgVCqrvC995Kc6c6MLmUyGuLg4+Pj4VLqNmpyXJ8+PxYsX4/r164iJiTHIvkpTk3PzpJiYGKxbtw5nzpwpsezkyZMICgrCzZs3K92GrmpCH/S0PqWi9N0HlUZ0zvr06YPXXnsNISEheomDqpeF6ACopH5r+1Vp+z0T9+i03sCBA7F3717cuXMHLi4uVdpnWerXr6/3IfmP5mys9Lbvho8td52AgACcOXMGtWrVglwuh5eXFwYMGIAZM2bA1ta23O2DgoLg7u6OJUuWVDrOyji3fEilt233zo6nLg8ICMCIESPg4+ODESNGICkpqVL7CQsLq9R2VRUb+3+V3tbf/7Myl9nY2Gj+/+DBA9SqVQsWFo+71bCwML0db6dOncotwvSp1ovzqrT9w9PvP3V5aYXWsWPHqnRulcUQfVBpDv9ctX67e5uy++3evXujRYsW+Oijj7TmX716Fa1bt8bt27fh6emp0370WeSSfvDSpJlKTk7GDz/8AFtbW2zevLnM9R49elSNUdUcK1asQE5ODjIyMrBx40acPXsWL730Eu7fvy86NKpBcnNzNV/PP/881q1bp5muSBFmrj9nVSVJEgoLC0WHYXDBwcH48ssvSxxrTEwMXn75ZZ2LMKqZWIiZqS+//BKNGzfGu+++i+joaM38hQsXYvDgwRg3bhzs7e3x0Ucf4Y8//kD37t3h5OQEZ2dnBAYGIiMjQ6u9n3/+GS1atIBKpcKQIUOQmZkJAEhISIBMJsM///wDAMjKysIbb7yBZ555BiqVCp06dcKDBw+q7bgrSi6Xo02bNti+fTuSk5MRHR2NhQsXYsSIEZp1/vnnH8hkMiQkJGDt2rXYvHkzIiIiYGNjg86dOwMo/7hPnDgBHx8fqFQqjB49GgUFBdV+rOXJz89Hnz59kJycDBsbG9jY2CAuLk6n86NI8dwFBgbCw8NDk5MrV65olgUFBWHixIkYPHgwbG1t0bJlS1y6dMnQh6kXMTExeOGFF7Tmubu749ixYwBK/zkrbsmSJWjZsiXu3r2LY8eOwd3dXbNs2bJlqFevHmxtbeHt7Y1t27ZpbTtnzhw4OTmhbt26T/1Dyxjcvn0bvr6+WLlyJYDHo7NhYWEICAiAtbU1zp07h3379qFt27aws7NDvXr1MG/evyN6xfuggIAAzJs3D127doWtrS1eeukl/Pnnn0KOTVcDBw5Efn4+Dhw4oJlXWFiIzZs3Izg4GAEBAVi3bp1m2ZdffonmzZvD3t4enTp1wq+//grg8c/b7du3MWjQINjY2GDOnDkAHt8e8dlnn5XaB2VnZ6N///5wdXWFg4MD+vbti9u3b2v2FRAQgDlz5qBLly6wtrZGQEAAkpOT8c4778DR0RHe3t6a855Kx0LMTEVHR2P06NEYNWoUrly5gosXL2qWff/993j55ZeRnp6OKVOmQJIkzJw5E3fv3sW1a9eQmJio1dEBj3/x7Nq1C3///Tfy8/Px9ttvl7rfsWPH4t69e/j555+Rnp6OJUuWQC6v+aehSqVCz549ceLEiaeuN3HiRIwaNQrTpk1Dbm6uZv3yjvubb77BqVOn8Pvvv+P06dP48ssvDXo8laFUKrFv3z64urpqRn18fX11Oj/K8vLLLyM+Ph7Jyclo164dAgMDtZZv2bIF06ZNQ2ZmJrp161bmeWWMiv+cFZEkCdOmTcPu3btx/PjxEqMd8fHxmD9/Pn788Ufk5OTg1KlTaNmypWb5hQsX4O7ujnv37mH16tWYMGECsrOzq+uw9OrXX39Fly5dEBYWhtDQUM38mJgYrFixArm5uWjbti2sra0RExODzMxM7N27F1FRUfjmm2/KbHfTpk1YvXo10tLS4OHhofP5KopSqURgYKDWvZX79+9HQUEBXn31Va11v//+e8ydOxdbt25FWloaRo8ejf79+6OgoABbt25F/fr18e233yI3Nxfh4eGa7crqg9RqNcaOHYuEhAT89ddfsLOzw8SJE7X2uXnzZkRGRiIlJQWFhYV44YUX0KxZM6SkpGDixIn4v/+r/K0B5oD3iNVAD9INO0IUGxuLX3/9FaNGjUK9evXQuXNnREdHo23btgAAPz8/zaiFlZUVvL294e3tDeBxhzB16lTMnTtXq8233npLs054eDjatWtX4obspKQk7Nq1C8nJyZp70l566aUKx5+RlVbhbfTB09NTq2DVlS7HHRYWBicnJwDAK6+8gosXL2LcuHEV20+GmMumupwfZQkODtb8f/78+YiIiEBaWpomFwMHDkTHjh0BAGPGjMH69esrFWNSUs17dEDxnzPg8SXKsWPHIjk5GYcOHYK1tXWJ7SwsLCBJEq5evYr69evDw8MDHh4emuV169bF5MmTAQCDBw9GUFAQrl+/XqEbyqX8rKocml6cPn0aISEhiIyMRL9+2vdfjRkzBq1btwbw+JwrGnkGgBYtWiAwMBDHjx/Ha6+9VmrbwcHBaN68OQBg5MiRmD9/fpXjvVfLsP12cHAwOnbsiIyMDDg4OGDjxo0YOXJkiQ8gfPrpp5g5cyZatGgBAJgwYQKWLl2KM2fOaOWpuLL6IHt7e608zp49W/MzWSQoKEjzYaNBgwbhs88+0/RfI0eOxLvvvovc3FyteyrpXyzEaqDDcw8btP3o6Gh07NgRDRo0AAC8/vrrmDFjBpYtWwYA8PLy0lr/3r17CA0NxcmTJ5GTkwO1Wl3ipvX69etr/u/l5YWCggKkpKRorXP79m2oVKoqfzBg8SfTqrR9Zd29exeOjo4V3k6X437yslOdOnUqdcPyqx/sq/A2+qDL+VGawsJCzJkzB9u3b0dKSopmhDA1NVXzC6F4XvLy8ioVY79+Wyq1nSEV/zkDgFu3buHKlSs4duxYqUUYADRq1AgbN27EqlWrEBwcjE6dOmH58uWaX4RP5gx4nLeK3qz+6OKqCq1fUbVq1cLDhw+15j18+BC1atXSTK9btw5+fn7o27dvie2L5+7s2bOYNWsWrl69ioKCAuTn52PQoEFl7r/4eaWPm/lHNjdsv+3v748mTZpg27ZtGDFiBHbv3o3//e9/JdZLSEjAu+++i9mzZ2vmFRQU4M6dO09tv6w+6P79+5g6dSr279+vueUgNzcX+fn5miKw+LbFp4u2YSFWOhZiNVD3D7pXafunFXL5+fnYunUr8vPzNT8sjx49Qnp6Onbv3g3g8f0CTwoLC4Narcbly5fh5OSEXbt2YcKECVrrPHnPwO3bt1GrVi24uLhoza9fvz6ysrKQmpoKZ2fnSh9f2FsRld62skVcdnY2fvzxR8ydOxcPHjzQumm/eNFUPH/6Ou7y7J7bp9Lb6lrEFT82QLfzozRbtmzBzp07cejQITRs2BDZ2dmwt7eHIZ6os2fPyEpvW5kizsbGRuscefjwIdLT07XWKS2Xzz77LGbMmIEBAwZg37598PPzK7X9YcOGYdiwYbh//z5mzZqF8ePH4+TJkxWOsywWbat2Cbi8Qq5+/fr4448/tObdunVLq8Bau3YtVq5ciXHjxmHDhg1al/KL527kyJF48803sXfvXlhZWWHq1KlITEys0jFU1JarVeu3dSnkgoODsXHj40+NN23aVHMV40n169fHjBkzEBQUVGobpZ13T7N8+XL89ttvOHPmDDw8PPDLL7+gdevWBvk5NVcsxGogK0crg7X93XffoaCgAFeuXNH8pQIA06dPR3R0NNq1a1dim5ycHFhbW0OlUuHu3buakbMnrV27Fv369YOzszPmzp2L4cOHl7j3y93dHf3798cbb7yBdevWwcHBAWfPnoWfn1+Fnu/joHKqwBFXjSRJuHz5MmbNmgUnJycEBwfj9OnTCA8Px61bt+Dm5oZFixZpbePm5oZbt25ppvV13OVxd6hT/kpV5ObmhoyMDM3lEUC386M0OTk5UCqVcHJywoMHD3S+nFkZ7u7V+5d4q1atcO3aNZw7dw6tW7fGggULoFarddp28ODBkCQJffv21dyE/qT4+Hj89ddf6NixI5RKJWxsbKBQKPQav0yp0mt7xY0aNQrh4eHo1q0bmjVrhmvXriEiIkLrPjBra2vs3bsXffv2RUhICD7//PMy7yfNycmBg4MDrKysEBsbiy1btqBr164GPYbi3B4art8uMnr0aMyaNQtJSUla9xU+6c0338TMmTPRpk0btGzZEnl5eTh69CgCAgJga2sLNze3Cj0KJScnB1ZWVrC3t0dGRgbef//pjyahiqv5d0mTXkVHR2PMmDHw9vaGu7u75mvq1Kk4ePAg7t69W2KbBQsW4PLly7C3t0ffvn0xcODAEuuMGTMGr776Kp555hkoFArNJ5yK27hxI1QqFVq0aAEnJyfMnj1b519Q1WnKlCmwtbWFvb09Xn/9dbRt2xanTp1CnTp10L17d4wZMwZ+fn5o3rw5evToobVtSEgIfv/9dzg4OGh+GRjLcZfHx8cHo0ePRuPGjWFvb4+4uDidzo/SFJ2HdevWha+vL55//nnDBl+NmjRpgg8++ACvvPIKvLy8UK9evQqNhg4ZMgRr1qxBnz59StyXmJ+fjzlz5sDFxQVOTk44c+aM1ifmjEFwcDAmTZqEIUOGwM7ODgMGDMC4cePwxhtvaK1nY2ODvXv34saNG/jPf/5T5s/M2rVr8d5778HW1hYLFy7E0KFDKx1bs2bNauwnTV1dXdG3b1/cvXsXo0aNKnWdAQMGYMGCBRg7dizs7e3RpEkTfPHFF5rls2fPxtKlS2Fvb6/ThxSmTJmCgoICuLi4oH379ujZs2eVjuGNN94o8X02d3yyfg3j7++vlwe61rSnx+uLv79/lR/oaoq58ff3r/IDXU0xL8Dj3FT1ga6mmpvS+Pv76+WBruaWM3080NWcckb/4qXJGkjXJ+ObK12ejm+Oyns6vjl72tPxqaTynoxPJT3tyfhET8MRMSOSlJSEfv36Yc+ePSU+GUXMT1mYl7IxNxXDfFUcc0bl4T1iRERERIKwECMiIiIShIWYEbGxscH48eP5ULwyMD+lY17KxtxUDPNVccwZlYf3iBEREREJwhExIiIiIkFYiBEREREJwkKMiIiISBCzeaBramqq6BD0SiaTwcrKCg8ePODLV/WMuTUs5tdwmFvDYn4Nx1Rzq8urzTgiZqTkcjnq1KlT5ktwqfKYW8Nifg2HuTUs5tdwzDm35nfERERERDWEURdiS5YsgUwmw5QpU0SHQkRERFRhRluInT9/HpGRkWjZsqXoUIiIiIgqxSgLsdzcXIwaNQpRUVFwcHAQHQ4RERFRpRhlITZp0iS88sor6NGjh+hQiIiIiCrN6B5fsW3bNly8eBHnz59/6nqJiYlITEzUTCuVSnh6eho6vGqjUCi0/iX9YW4Ni/k1HObWsJhfwzHn3BpVIfbXX38hNDQUhw4dQu3atZ+6bmRkJBYtWqSZDgsLQ3h4uKFDrDa5ubmIjIzEqFGjYGdnJzock8S8GhbzazjMrWExv4Zjjrk1qpd+f/fddxg0aJBWxVxYWAiZTAa5XI78/HzNMlMfEUtNTUVgYCC2bt2q0wPjSHcKhQJ2dnbIzs5GYWGh6HBMDvNrOMytYTG/hmOqudXlPnajGhHr3r07rly5ojUvODgYPj4+mDlzplaB5uHhAQ8PD810amqqSX1z1Wq15l9TOq6apLCwkLk1IObXcJhbw2J+Dcccc2tUhZitrS2aN2+uNc/a2hpOTk4l5hMRERHVdEb5qUkiIiIiU2BUI2KlOXbsmOgQiIiIiCqFI2JEREREgrAQIyIiIhKEhRgRERGRICzEiIiIiARhIUZEREQkCAsxIiIiIkFYiBEREREJwkKMiIiISBAWYkRERESCsBAjIiIiEoSFGBEREZEgLMSIiIiIBGEhRkRERCQICzEiIiIiQViIEREREQnCQoyIiIhIEBZiRERERIJYiA6AqDS//vqrsH3L5XLY2toiJycHarVaWBzNmjUTtm8iIqoeLMSoRnrnnXdEhyDcV199BZVKJToMIiIyILMpxCwtLaFUKkWHoTf3798HAFhZWcHW1lZwNPqVmZkpOoQaITAwEN9//z3s7e1Fh6JXeXl5iIyMxNChQ03u3BVNJpMBAKytrSFJkuBoTA/zazjmnFuzKcQKCgpQUFAgOgy9efDggebfnJwcwdHoV1paGgBgcJ8gWNcxz1/UefdzsHNfDNLS0qBQKESHo1dpaWmIiopCQECA0Eu/pkihUMDS0hJ5eXkoLCwUHY7JYX4Nx1Rzq8sAkNkUYmR8fBq1hIPKSXQYQmRkpYkOgYiIqgE/NUlEREQkCAsxIiIiIkFYiBEREREJwkKMiIiISBAWYkRERESCsBAjIiIiEoSFGBEREZEgLMSIiIiIBGEhRkRERCQIn6xPZGaysrKQn58vbP8ZGRkAgNTUVKGvMlEqlXypOhEJx0KMyIxkZWUhMHAE1GrxL9UNDQ0Vun+5XIatW7exGCMioViIEZmR/Px8qNUS1k8OgKvKSnQ4wiRnPcB/Vh8TOjJIRASwECMyS64qK7g71BEdBhGR2ePN+kRERESCsBAjIiIiEoSFGBEREZEgLMSIiIiIBDG6QuzDDz/E888/D1tbW7i6umLgwIGIj48XHRYRERFRhRldIXb8+HFMmjQJZ86cwaFDh/Dw4UO8/PLLyMvLEx0aERERUYUY3eMr9u/frzUdExMDV1dXXLhwAZ07dxYUFREREVHFGV0hVlxWVhYAwNHRUXAkRMbjt78ycC/zvugwhEnL4YNciahmMOpCTK1WY8qUKXjppZfQvHlzrWWJiYlITEzUTCuVSnh6eupt36Lf15eenq71ryiGeF+fQqEAAGTlZOi1XWNSdOwKhUKTD30oamvWxjN6a9OY6Tu/ohUdiykdU03C/BqOOefWqAuxSZMm4erVq/jpp59KLIuMjMSiRYs002FhYQgPD9fLfjMzMzF02FBINeB9fZMnTxa6f5lchkMHD8He3l5/bcpkkMvlWLPxfb21aYzkcjnc3Nz0mlu+0kebSqWCg4OD6DD0zs7OTnQIJik3NxeRkZEYNWoUc6xn5pxboy3E3nrrLezZswcnTpzAM888U2L5hAkT8Oqrr2qmlUolMjL0M8KSnJwMSS3hpekvobZ9bb20aYz+yfwHp5adwr179yBJ+i1Kv/76a2FFQ3p6OiZPnozVq1cLveStVCohSZLezlvg30v5S8a+ACdbpd7aNTZpOfmYtfEMsrKyoFSaTh4UCgXs7OyQnZ2NwsJC0eGYnNTUVERFRaFbt25wdnYWHY5JMdXc6vKHntEVYpIkYfLkyfj2229x7NgxNGzYsNT1PDw84OHhoZlOTU3VW8dU1E5t+9qwcjTfFycXKSws1Hunb2NjAxsbG722WVGOjo5wcnISGoO+81rU3nP1HMz6XZNJGY/vjzPEuVsTmOpxiaZWqzX/Mr/6Zc65NbpCbNKkSdiyZQt27doFW1tbJCUlAXh8icHKikURERERGQ+je47Yp59+iqysLAQEBGhGvTw8PPDVV1+JDo2IiIioQoxuREzf9yIRERERiWJ0I2JEREREpoKFGBEREZEgLMSIiIiIBGEhRkRERCQICzEiIiIiQViIEREREQlidI+vIKKqS856IDoEocz9+IlKk5WVJezVckWvctPnW3AqQ6lUQqVSVes+WYgRmRGlUgm5XIb/rD4mOhTh5HKZSb1nkqgqsrKyEDhiBNSCn9UZGhoqdP9ymQxbt22r1mKMhRiRGVGpVNi6dZuwv3qBx3/5hoaGYuXKlTq9ENdQRPzlS1RT5efnQy1JWBX/Epwf1hYdjhCptf7B201PVXv/yEKMyMyILj4UCgUAwNnZWfhL1YlIm/PD2nB7yPc2VyferE9EREQkCAsxIiIiIkFYiBEREREJwkKMiIiISBAWYkRERESCsBAjIiIiEoSFGBEREZEgOj9HLDc3F5cvX4ZcLscLL7wAABg5cmSp665evZrPByIiIjIy8XUykfLIPF8Blm4h5kHXOhdiMTExCA0NxaBBg/DNN98AALZt2waZTFZi3fbt2wt/TQERERFVzCLvC6JDMDs6F2Lff/89AGDMmDEllklPvJtKJpPhwIEDNa4Qs7S01Nt75fLy8vTSjqmwtraGra2t6DD0Ri6XY/z48XBxcYG1tbXocEwO82s4eXl5iIyMxNChQ03qZ7KmuH//PgDAysrK5PIr8kXbNYlcJoOTk1O1fn91LsR+//13AMCLL75YYtkff/wBALhx4wZ69uyJK1eu6Ck8/SkoKEBBQYFe2mIhpi0vLw85OTmiw9AbhUKBCRMmICMjw6SOq6Zgfg0nLS0NUVFRCAgIgFqtFh2OyXnw4IHmX1M7dxUKBRYtWoSHDx8K2b9cLoeNjQ1yc3OFnrstWrSAQqHQ2/dXlwEgnQuxpKQkAICjo6Nm3rPPPgu5XA4vLy8AgKenJwAgJSWlQoESERGRWO3btxe2b4VCAQcHB2RkZJjd6JzOhVitWrWQn5+PP//8E97e3gCAa9euaa2TnJz8uFELvkuciIiIqDw6P76iqPj6+OOPy1znk08+AQDNCBkRERERlU3nQqxPnz6QJAlr167FiBEjcO7cOWRlZSE7OxuxsbEYO3Ysli5dCplMhj59+hgyZiIiIiKToPM1xClTpiAqKgrp6enYvn07tm/fXmIdSZKgUqkwdepUvQZJREREZIp0HhFzdXXFjh07oFKpIElSqV8qlQo7duxA3bp1DRkzERERkUmo0CuOOnfujKtXr2LKlCnw8fGBlZUVrKys4OPjgylTpuDq1avo1q2boWIlIiIiMikV/nijp6cnIiIiEBERYYh4iIiIiMxGhQqxhw8fYsuWLYiNjQUA+Pv7Y8SIEXp7Yj0RERGROdG5EMvKykKXLl1KPDV/2bJlOHHiBBwcHPQeHBEREZEp0/kesYULF+Ly5cslbtD/7bffsHDhQgOGSERERGSadC7Edu/eDZlMBh8fH0RERGD58uVo2rQpJEnSvBCciIiIiHSn86XJv/76CwDwww8/oGHDhgCAV199FU2aNMGdO3cMEx0RERGRCdN5ROzRo0cAoCnCAKBRo0Zay4iIiIhIdxV+fMWmTZt0mj9mzJjKRURERERkJipciAUHB2tNy2SyEvNlMhkLMSIiIqJyVKgQkyTJUHEQERERmR2dC7EFCxYYMg4iIiIis1PhQuz48eOaJ+v7+fkhICDAIIGVZ82aNfjoo4+QlJSEVq1aYfXq1WjXrp2QWIiIiIgqQ+dCLD8/HwMHDsTBgwe15nfr1g3ff/89ateurffgyvLVV19h2rRpWLduHdq3b48VK1agV69eiI+Ph6ura7XFQURERFQVOj++YsmSJThw4ECJJ+sfOXIEixcvNmSMJURERGD8+PEIDg7Gc889h3Xr1qFOnTrYsGFDtcZBREREVBU6j4ht27YNAODq6ooRI0ZAkiRs27YNKSkp+Prrr/Hee+8ZLMgnFRQU4MKFC5g9e7ZmnlwuR48ePXD69OlqiaFI5p+ZeJDxoFr3WZPkZ+eLDoGIiMio6VyI/fHHH5DJZDhw4ABatWoF4PEjK9q2bYs///zTYAEWl5qaisLCQri5uWnNd3Nzw7Vr1zTTiYmJSExM1EwrlUp4enrqJQaFQgEAuBB1QS/tGTuFQqHJiSkoOhZTOqaaxNTzm5WVhfx8MX+kpKena/0rilKphEql0nu7InMLmH5+RTL1fuFpdC7ECgoKIJPJNEUYALRu3VqzrKaJjIzEokWLNNNhYWEIDw/XS9siO4KaSKVSwcHBQXQYemdnZyc6BJNmivnNzMzEsGHDoFarhcYxefJkofuXy+U4ePAg7O3t9dZmZmYmhg4dBkkSm1tAfH5lMjkOHdJvfmsKU+wXylPhB7qePHmy1OeJFZ/fuXPnqkVWBmdnZygUCty7d09r/r179+Du7q6ZnjBhAl599VXNtFKpREZGhl5iyMrKAgD4jfeD0k6plzaNUX52Pi5EXUBWVhaUStPJg0KhgJ2dHbKzs1FYWCg6HJNjyvlNTk6GWq3Ghg0D4OpqLTocIZKT8zBu3C7cu3dPr8+eTE5OhiSpoWgeDJmlrd7aNTZSQQ4Kr0brPb+imWq/oMsgRYULseKPqyh6sv6T82UymcHeP2lpaQk/Pz8cPnwYAwcOBACo1WocPnwYb731lmY9Dw8PeHh4aKaLLmnqQ1E79l72sHK00kubxuhB+uP74woLC03qB6eIqR5XTWGK+S06HldXa7i72wiORix9f3+L2pJZ2kKmNK3LcpVhij8/gOke19MY5ZP1p02bhrFjx8Lf3x/t2rXDihUrkJeXV+L1S0REREQ1mc6F2NixYw0ZR4UMHz4cKSkpmD9/PpKSktC6dWvs37+/xA38RERERDWZzoVYdHS0IeOosLfeekvrUiQRERGRsdH5ga5EREREpF8sxIiIiIgEYSFGREREJAgLMSIiIiJBWIgRERERCcJCjIiIiEgQFmJEREREgrAQIyIiIhKEhRgRERGRICzEiIiIiARhIUZEREQkCAsxIiIiIkFYiBEREREJwkKMiIiISBAWYkRERESCsBAjIiIiEoSFGBEREZEgFqIDICIyNb/9loKkpFzRYQiRnv5AdAhERoWFGBGRns2YcUh0CERkJMymELO0tIRSqdRLW3l5eXppx1RYW1vD1tZWdBh6I5PJADw+LkmSBEdjekw5v+wb/qXvfoG51cZ+13SYTSFWUFCAgoICvbTFDkFbXl4ecnJyRIehNwqFApaWlsjLy0NhYaHocEyOKee3qG9YurQnHB2tBEcjRnr6A8yYcUjv/QL7XW3sd42DLgNAZlOIERFVl+eec4G7u43oMIQw13vjiCqLn5okIiIiEoSFGBEREZEgLMSIiIiIBGEhRkRERCQICzEiIiIiQViIEREREQnCQoyIiIhIEBZiRERERIKwECMiIiIShE/Wr4J/Mv8RHYJQ5n78REREVcVCrBKUSiVkchlOLTslOhThZHKZ3l6mTkREZG5YiFWCSqXCtq3bkJ+fLyyGjIwMhIaGYuXKlXBwcBAWh1KphEqlErZ/IiIiY8ZCrJJEFx8KhQIA4OzsDCcnJ6GxEBERUeXwZn0iIiIiQViIEREREQnCQoyIiIhIEBZiRERERIIYVSGWkJCAkJAQNGzYEFZWVmjUqBEWLFiAgoIC0aERERERVZhRfWry2rVrUKvViIyMROPGjXH16lWMHz8eeXl5WLZsmejwiIiIiCrEqAqx3r17o3fv3pppb29vxMfH49NPP2UhRkREREbHqAqx0mRlZcHR0VF0GEREVA2k3LuQ8rNFhyHOw1zREZCeGXUhduPGDaxevbrU0bDExEQkJiZqppVKJTw9PaszPIOSy+Waf4se7kr6UZRP5tUwTDm/RceUnJwnOBJxio5doVDo9Xtc1Fbh9W/01qYx03d+RTPlfqE8NaIQmzVrFv773/8+dZ24uDj4+Phopu/cuYPevXtj6NChGD9+fIn1IyMjsWjRIs10WFgYwsPD9Re0YEWvV7K1tRX6iiNTZmdnJzoEk2aK+ZXJZJDL5Rg3bpfoUISSy+Vwc3ODvb293tqUyWR6a8vYyWT6z29NYYr9QnlkkiRJooNISUlBWlraU9fx9vaGpaUlAODu3bsICAjACy+8gJiYGM3o0JNMfUQsNTUVgYGB2Lp1K5ydnUWHY1IUCgXs7OyQnZ2NwsJC0eGYHFPPb1ZWlrD30Kanp2Py5MlYvXq10Fs2DPUO2rNnzwr9lLxcLkedOnVw//59qNVqYXG0bNlS+Gv29M1U+wVdBkpqxIiYi4sLXFxcdFr3zp076Nq1K/z8/BAdHV1qEQYAHh4e8PDw0Eynpqaa1De3qBNQq9UmdVw1SWFhIXNrQKaaXxsbG9jY2AiNwdHRUfg7aA3xvfX399d7mxWhUCjg4OCAjIwM4eeu6P0biqn2C09TIwoxXd25cwcBAQHw8vLCsmXLkJKSolnm7u4uMDIiIiKiijOqQuzQoUO4ceMGbty4gWeeeUZrWQ24wkpERERUIUb1ZP2goCBIklTqFxEREZGxMapCjIiIiMiUsBAjIiIiEoSFGBEREZEgLMSIiIiIBGEhRkRERCQICzEiIiIiQViIEREREQnCQoyIiIhIEBZiRERERIKwECMiIiIShIUYERERkSAsxIiIiIgEYSFGREREJAgLMSIiIiJBWIgRERERCcJCjIiIiEgQFmJEREREgrAQIyIiIhKEhRgRERGRIBaiA6gulpaWUCqVosPQG7lcjvHjx8PFxQXW1taiwzEpMpkMAGBtbQ1JkgRHY3qYX8Nhv2BYPHcNx5xzK5PM5IhTU1NFh6BXCoUCDg4OyMjIQGFhoehwTApza1jMr+Ewt4bF/BqOqebW2dm53HV4aZKIiIhIEBZiRERERIKwECMiIiIShIUYERERkSAsxIiIiIgEYSFGREREJAgLMSIiIiJBzOY5YqYmMTERkZGRmDBhAjw8PESHY1KYW8Nifg2HuTUs5tdwzDm3HBEzUomJiVi0aBESExNFh2JymFvDYn4Nh7k1LObXcMw5tyzEiIiIiARhIUZEREQkCAsxI+Xh4YEFCxaY3bX06sDcGhbzazjMrWExv4ZjzrnlzfpEREREgnBEjIiIiEgQFmImoEGDBlixYsVT15HJZPjuu++qJR5zlJCQAJlMhkuXLokOpcYKCgrCwIEDNdMBAQGYMmWKsHhMTXn9QEXyberfm+Lnor6YSz+wcOFCtG7dWnQYJoOFWDU7ffo0FAoFXnnllWrdb2JiIvr06VOt+6xuQUFBkMlkkMlkqFWrFtzc3NCzZ09s2LABarVa53aOHTsGmUyGzMxMwwVrZJ78xaWvX9I7d+7E+++/X+V2jE3ROVrW18KFCw2yX3PJd1nnZ0xMDOzt7Q2673r16iExMRHNmzc36H4MqX///ujdu3epy06ePAmZTIbBgwfj8OHDOrdpLgVqZbEQq2aff/45Jk+ejBMnTuDu3btlridJEh49eqS3/bq7u0OpVOqtvZqqd+/eSExMREJCAvbt24euXbsiNDQU/fr102s+qeocHR1ha2srOoxql5iYqPlasWIF7OzstOZNnz69Qu0VFBTotJ655lsfdM2xQqGAu7s7LCwsDByR4YSEhODQoUP4+++/SyyLjo6Gv78/WrZsCScnJwHRmSYWYtUoNzcXX331Fd5880288soriImJ0SwrGoXZt28f/Pz8oFQq8dNPP+HmzZsYMGAA3NzcYGNjg+effx4//vhjibZzcnIQGBgIa2tr1K1bF2vWrNFaXvzS5N9//43AwEA4OjrC2toa/v7+OHv2rKEOvdoolUq4u7ujbt26aNu2LcLCwrBr1y7s27cPMTExpf5llpmZCZlMhmPHjiEhIQFdu3YFADg4OEAmkyEoKAgAoFarsXTpUjRu3BhKpRL169dHeHi41v5v3bqFrl27ok6dOmjVqhVOnz5dXYdeLYKCgnD8+HGsXLlSM4KTkJCAwsJChISEoGHDhrCyskLTpk2xcuXKp7ZVfOTiiy++gL+/P2xtbeHu7o6RI0ciOTlZs7zoZ+Tw4cPw9/dHnTp10KFDB8THxxvqcA3C3d1d86VSqSCTyTTT69atQ8eOHbXWX7FiBRo0aKCZLhqdDA8Ph6enJ5o2bVrqftavXw97e3vNyEXxfK9duxZNmjRB7dq14ebmhtdee01re7VajRkzZsDR0RHu7u4GG6mrCc6fPw8XFxf897//BfDvpbf169ejYcOGqF27NgBg//796NixI+zt7eHk5IR+/frh5s2bmnaK9y/GeM7269cPLi4uWr+fgMe/v7Zv346QkJBSL02uX78evr6+qF27Nnx8fLB27VrNsoYNGwIA2rRpA5lMhoCAAAD/nsvLli2Dh4cHnJycMGnSJDx8+FCzra79woEDB9CmTRtYWVmhW7duSE5Oxr59++Dr6ws7OzuMHDkS9+/f12+y9ISFWDX6+uuv4ePjg6ZNm2L06NHYsGEDin9oddasWViyZAni4uLQsmVL5Obmom/fvjh8+DB+/vln9O7dG/3798ft27e1tvvoo4/QqlUr/Pzzz5g1axZCQ0Nx6NChUuPIzc1Fly5dcOfOHezevRu//PILZsyYUaHLd8akW7duaNWqFXbu3FnuuvXq1cOOHTsAAPHx8UhMTNQUFLNnz8aSJUswb948/Pbbb9iyZQvc3Ny0tp8zZw6mT5+OS5cu4dlnn0VgYKBJjcStXLkSL774IsaPH68ZwalXrx7UajWeeeYZbN++Hb/99hvmz5+PsLAwfP311zq3/fDhQ7z//vv45Zdf8N133yEhIUFTBD9pzpw5WL58OWJjY2FhYYFx48bp8QiNw+HDhxEfH49Dhw5hz549JZYvXboUs2bNwsGDB9G9e/cSy2NjY/H222/jvffeQ3x8PPbv34/OnTtrrbNx40ZYW1vj7NmzWLp0Kd57770y+xRjduTIEfTs2RPh4eGYOXOmZv6NGzewY8cO7Ny5U1NY5eXlYdq0aYiNjcXhw4chl8sxaNCgcvtOYzpnLSwsMGbMGMTExGj9ftq+fTsKCwsRGBhYYpvNmzdj/vz5CA8PR1xcHBYvXox58+Zh48aNAIBz584BAH788UckJiZq9cVHjx7FzZs3cfToUWzcuBExMTFaRaCu/cLChQvxySef4H//+x/++usvDBs2DCtWrMCWLVvwww8/4ODBg1i9erWesqRnElWbDh06SCtWrJAkSZIePnwoOTs7S0ePHpUkSZKOHj0qAZC+++67cttp1qyZtHr1as20l5eX1Lt3b611hg8fLvXp00czDUD69ttvJUmSpMjISMnW1lZKS0ur4hHVLGPHjpUGDBhQ6rLhw4dLvr6+0h9//CEBkH7++WfNsoyMDAlAie9FRkaGZp3s7GxJqVRKUVFRpbZf1O769es183799VcJgBQXF1fVQxPuydx26dJFCg0NLXebSZMmSUOGDCm1DV3aOX/+vARAysnJkSTp3+/Ljz/+qFnnhx9+kABIDx48qNDx1BTR0dGSSqXSTC9YsEBq1aqV1joff/yx5OXlpZkeO3as5ObmJuXn52ut5+XlJX388cfSjBkzJA8PD+nq1atay5/M944dOyQ7OzspOzu71Li6dOkidezYUWve888/L82cObNiByhAWefVk7kuOhd37twp2djYSNu2bdNad8GCBVKtWrWk5OTkp+4rJSVFAiBduXJFkiSpRP9irOdsXFycVp8oSZLUqVMnafTo0ZIklTxPGzVqJG3ZskWrjffff1968cUXJUkqmZciY8eOlby8vKRHjx5p5g0dOlQaPnx4mbHp0i98+OGHEgDp5s2bmnkTJkyQevXqpVsCqhlHxKpJfHw8zp07p/lrwsLCAsOHD8fnn3+utZ6/v7/WdG5uLqZPnw5fX1/Y29vDxsYGcXFxJUbEXnzxxRLTcXFxpcZy6dIltGnTBo6OjlU9LKMhSRJkMlmlt4+Li0N+fn6powtPatmypeb/RQ8mfHIY3ZStWbMGfn5+cHFxgY2NDT777LMS5+nTXLhwAf3790f9+vVha2uLLl26AECJNsw5x0VatGgBS0vLEvOXL1+OqKgo/PTTT2jWrFmZ2/fs2RNeXl7w9vbG66+/js2bN5e4bPNknoHHuTalPJ89exZDhw7FF198geHDh5dY7uXlBRcXF615v//+OwIDA+Ht7Q07OzvNJePyznNjO2d9fHzQoUMHbNiwAcDj0cGTJ08iJCSkxLp5eXm4efMmQkJCYGNjo/n64IMPtC7blqVZs2ZQKBSa6eLnWWX6BTc3N9SpUwfe3t5a82pqzlmIVZPPP/8cjx49gqenJywsLGBhYYFPP/0UO3bsQFZWlmY9a2trre2mT5+Ob7/9FosXL8bJkydx6dIltGjRQuebR0tjZWVV6W2NVVxcHBo2bAi5/PEpLz0x5P7k/Qhl0TVntWrV0vy/qPAz1Uu+T9q2bRumT5+OkJAQHDx4EJcuXUJwcLDO52leXh569eoFOzs7bN68GefPn8e3334LoOSN0qacY7lcXuJ2hdLOz+L9RJFOnTqhsLCw3EvCtra2uHjxIrZu3QoPDw/Mnz8frVq10vqk8JN5Bh7n2hjybGdnp9WnFsnMzIRKpdJMN2rUCD4+PtiwYYPOOe7fvz/S09MRFRWFs2fPau6rLe88N8ZzNiQkBDt27EBOTg6io6PRqFEjTRH0pNzcXABAVFQULl26pPm6evUqzpw5U+5+nnaeVaVfMKbzl4VYNXj06BE2bdqE5cuXa52ov/zyCzw9PbF169Yytz116hSCgoIwaNAgtGjRAu7u7khISCixXvET/syZM/D19S21zZYtW+LSpUtIT0+v0nEZiyNHjuDKlSsYMmSI5i/cxMREzfLiH6kuGmkoLCzUzGvSpAmsrKwq9JFtU2VpaamVG+DxedqhQwdMnDgRbdq0QePGjXX6a7jItWvXkJaWhiVLlqBTp07w8fGpsX+9GpKLiwuSkpK0irGKfOS/Xbt22LdvHxYvXoxly5Y9dV0LCwv06NEDS5cuxeXLl5GQkIAjR45UNvQao2nTprh48WKJ+RcvXsSzzz6rmXZ2dsaRI0dw48YNDBs2rNw/yNLS0hAfH4+5c+eie/fu8PX1RUZGht7jrymGDRsGuVyOLVu2YNOmTRg3blypVxXc3Nzg6emJW7duoXHjxlpfRTfpl9an6sJc+gUWYtVgz549yMjIQEhICJo3b671NWTIkBKXJ5/UpEkTzc2iv/zyC0aOHFlqVX/q1CksXboU169fx5o1a7B9+3aEhoaW2mZgYCDc3d0xcOBAnDp1Crdu3cKOHTtM4hN++fn5SEpKwp07d3Dx4kUsXrwYAwYMQL9+/TBmzBhYWVnhhRde0Hwg4vjx45g7d65WG15eXpDJZNizZw9SUlKQm5uL2rVrY+bMmZgxYwY2bdqEmzdv4syZM0/93pmqBg0a4OzZs0hISEBqairUajWaNGmC2NhYHDhwANevX8e8efNw/vx5ndusX78+LC0tsXr1aty6dQu7d+82i2deFRcQEICUlBQsXboUN2/exJo1a7Bv374KtdGhQwfs3bsXixYtKvMBr3v27MGqVatw6dIl/Pnnn9i0aRPUanWZn8A0Jm+++SauX7+Ot99+G5cvX0Z8fDwiIiKwdetWvPPOO1rrurq64siRI7h27Vq5H6xxcHCAk5MTPvvsM9y4cQNHjhzBtGnTqhzvuXPn4OPjgzt37lS5LX2ysbHB8OHDMXv2bCQmJpZ6g3yRRYsW4cMPP8SqVatw/fp1XLlyBdHR0YiIiADwOM9WVlbYv38/7t27V+qIZWkM2S988skn5d5qUl1YiFWDzz//HD169NAaFi8yZMgQxMbG4vLly6VuGxERAQcHB3To0AH9+/dHr1690LZt2xLrvfPOO4iNjUWbNm3wwQcfICIiAr169Sq1TUtLSxw8eBCurq7o27cvWrRogSVLlmhdpzdW+/fvh4eHBxo0aIDevXvj6NGjWLVqFXbt2qU5vg0bNuDRo0fw8/PDlClT8MEHH2i1UbduXSxatAizZs2Cm5sb3nrrLQDAvHnz8M4772D+/Pnw9fXF8OHDTfKvs/JMnz4dCoUCzz33HFxcXHD79m1MmDABgwcPxvDhw9G+fXukpaVh4sSJOrdZ9HH57du347nnnsOSJUvKHdExRb6+vli7di3WrFmDVq1a4dy5cxV+rhgAdOzYET/88APmzp1b6ifF7O3tsXPnTnTr1g2+vr5Yt24dtm7d+tT7yoyFt7c3Tpw4gWvXrqFHjx5o3749vv76a2zfvr3UB5W6u7trRs1HjRpV5qiNXC7Htm3bcOHCBTRv3hxTp07FRx99VOV479+/j/j4eJ1ukahuISEhyMjIQK9eveDp6Vnmev/5z3+wfv16REdHo0WLFujSpQtiYmI0I2IWFhZYtWoVIiMj4enpiQEDBui0f0P2C6mpqRUatTckvvSbiIiISBCOiBEREREJwkKMiIiISBAWYkRERESCsBAjIiIiEoSFGBEREZEgLMSIiIiIBGEhRkRERCQICzEiIiIiQViIEREREQnCQoyIiIhIEBZiRERERIKwECMiIiIS5P8Bu0MV1DbQWgQAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 700x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8737757232659)>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from plotnine import guide_legend, guides\n",
    "\n",
    "post[\"mt_rank_speed_diff\"] = post[\"mt_rank_speed_pe\"] - post[\"mt_rank_speed_ht\"]\n",
    "\n",
    "post_melted_temp = pd.melt(\n",
    "    post,\n",
    "    id_vars=['lang_id'],\n",
    "    value_vars=['mt_rank_speed_diff'],\n",
    "    var_name=\"modality\",\n",
    "    value_name=\"temporal_effort\"\n",
    ")\n",
    "post_melted_temp[\"lang_id\"] = post_melted_temp[\"lang_id\"].map(\n",
    "    {'ara': 'Arabic', 'ita': 'Italian', 'nld': 'Dutch', 'tur': 'Turkish', 'ukr': 'Ukrain.', 'vie': 'Vietnam.'}\n",
    ")\n",
    "post_melted_temp[\"modality\"] = post_melted_temp[\"modality\"].map(\n",
    "    {'mt_rank_speed_diff': 'Speed Increase'}\n",
    ")\n",
    "\n",
    "plot2 = (\n",
    "    ggplot(\n",
    "        post_melted_temp, aes(\"lang_id\", \"temporal_effort\", fill=\"lang_id\")\n",
    "    )\n",
    "    + geom_boxplot()\n",
    "    + xlab(\"Setting\")\n",
    "    + ylab(\"PPG\")\n",
    "    + theme(\n",
    "        legend_position=(0.50, 0.935),\n",
    "        legend_direction=\"horizontal\",\n",
    "        legend_box_spacing=0.05,\n",
    "        legend_key=element_blank(),\n",
    "        axis_text_x=element_text(colour=\"black\", size=10),\n",
    "        axis_text_y=element_text(colour=\"black\", size=10),\n",
    "        figure_size=(7,2.5),\n",
    "        axis_title_y=element_text(size=11, face=\"bold\", colour=\"black\"),\n",
    "        axis_title_x=element_blank(),\n",
    "        legend_title=element_blank(),\n",
    "        legend_box_margin=0.15,\n",
    "    )\n",
    "    + scale_fill_brewer(type=\"qual\", palette=\"Accent\", name=\"Language\")\n",
    "    + guides(fill = guide_legend(nrow=1))\n",
    ")\n",
    "plot2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gsarti/projects/venv/lib/python3.8/site-packages/plotnine/ggplot.py:820: PlotnineWarning: Filename: ../outputs/perceived_pe_speedup.pdf\n"
     ]
    }
   ],
   "source": [
    "from plotnine import save_as_pdf_pages\n",
    "\n",
    "save_as_pdf_pages([plot2], \"../outputs/perceived_pe_speedup.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>mt_rating</th>\n",
       "      <th>diff_mt</th>\n",
       "      <th>issues_order</th>\n",
       "      <th>issues_morph</th>\n",
       "      <th>issues_mistrans</th>\n",
       "      <th>issues_add</th>\n",
       "      <th>issues_term</th>\n",
       "      <th>issues_syntax</th>\n",
       "      <th>issues_verbs</th>\n",
       "      <th>issues_punct</th>\n",
       "      <th>...</th>\n",
       "      <th>mt_rank_speed_ht</th>\n",
       "      <th>mt_rank_speed_pe</th>\n",
       "      <th>mt_rank_cog_effort_ht</th>\n",
       "      <th>mt_rank_cog_effort_pe</th>\n",
       "      <th>mt_rank_tech_effort_ht</th>\n",
       "      <th>mt_rank_tech_effort_pe</th>\n",
       "      <th>mt_rank_pref_ht</th>\n",
       "      <th>mt_rank_pref_pe</th>\n",
       "      <th>duration_s</th>\n",
       "      <th>mt_rank_speed_diff</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mt_rating</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.192677</td>\n",
       "      <td>0.131085</td>\n",
       "      <td>-0.597348</td>\n",
       "      <td>-0.254082</td>\n",
       "      <td>-0.024940</td>\n",
       "      <td>-0.434990</td>\n",
       "      <td>-0.290389</td>\n",
       "      <td>-0.107789</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.340915</td>\n",
       "      <td>0.741651</td>\n",
       "      <td>0.047291</td>\n",
       "      <td>-0.091132</td>\n",
       "      <td>-0.083056</td>\n",
       "      <td>0.424611</td>\n",
       "      <td>0.067850</td>\n",
       "      <td>0.759264</td>\n",
       "      <td>-0.237797</td>\n",
       "      <td>0.693153</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>diff_mt</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>issues_order</th>\n",
       "      <td>-0.192677</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.419961</td>\n",
       "      <td>0.265317</td>\n",
       "      <td>-0.098844</td>\n",
       "      <td>-0.331890</td>\n",
       "      <td>0.651852</td>\n",
       "      <td>0.029654</td>\n",
       "      <td>0.300198</td>\n",
       "      <td>...</td>\n",
       "      <td>0.132865</td>\n",
       "      <td>-0.363105</td>\n",
       "      <td>0.077162</td>\n",
       "      <td>-0.040921</td>\n",
       "      <td>0.322302</td>\n",
       "      <td>-0.353757</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.440385</td>\n",
       "      <td>-0.298776</td>\n",
       "      <td>-0.320488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>issues_morph</th>\n",
       "      <td>0.131085</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.419961</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.175491</td>\n",
       "      <td>0.186797</td>\n",
       "      <td>-0.020907</td>\n",
       "      <td>0.309958</td>\n",
       "      <td>0.100874</td>\n",
       "      <td>0.363085</td>\n",
       "      <td>...</td>\n",
       "      <td>0.100437</td>\n",
       "      <td>-0.088226</td>\n",
       "      <td>0.174987</td>\n",
       "      <td>0.255202</td>\n",
       "      <td>-0.026104</td>\n",
       "      <td>0.025070</td>\n",
       "      <td>-0.206631</td>\n",
       "      <td>-0.105126</td>\n",
       "      <td>-0.354283</td>\n",
       "      <td>-0.127190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>issues_mistrans</th>\n",
       "      <td>-0.597348</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.265317</td>\n",
       "      <td>0.175491</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.219827</td>\n",
       "      <td>0.131844</td>\n",
       "      <td>0.379534</td>\n",
       "      <td>0.339471</td>\n",
       "      <td>0.043572</td>\n",
       "      <td>...</td>\n",
       "      <td>0.450564</td>\n",
       "      <td>-0.442500</td>\n",
       "      <td>-0.015152</td>\n",
       "      <td>0.344124</td>\n",
       "      <td>0.091474</td>\n",
       "      <td>-0.365617</td>\n",
       "      <td>0.007033</td>\n",
       "      <td>-0.497214</td>\n",
       "      <td>0.094627</td>\n",
       "      <td>-0.595383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>issues_add</th>\n",
       "      <td>-0.254082</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.098844</td>\n",
       "      <td>0.186797</td>\n",
       "      <td>0.219827</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.042454</td>\n",
       "      <td>0.165053</td>\n",
       "      <td>0.636229</td>\n",
       "      <td>0.102633</td>\n",
       "      <td>...</td>\n",
       "      <td>0.139054</td>\n",
       "      <td>-0.130292</td>\n",
       "      <td>0.050473</td>\n",
       "      <td>-0.032121</td>\n",
       "      <td>-0.195161</td>\n",
       "      <td>-0.219827</td>\n",
       "      <td>-0.017165</td>\n",
       "      <td>-0.094606</td>\n",
       "      <td>0.136314</td>\n",
       "      <td>-0.181125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>issues_term</th>\n",
       "      <td>-0.024940</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.331890</td>\n",
       "      <td>-0.020907</td>\n",
       "      <td>0.131844</td>\n",
       "      <td>0.042454</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.040641</td>\n",
       "      <td>0.156313</td>\n",
       "      <td>0.386810</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.104507</td>\n",
       "      <td>0.124597</td>\n",
       "      <td>0.382726</td>\n",
       "      <td>0.172409</td>\n",
       "      <td>0.301709</td>\n",
       "      <td>-0.163297</td>\n",
       "      <td>-0.199676</td>\n",
       "      <td>0.347242</td>\n",
       "      <td>-0.016808</td>\n",
       "      <td>0.151196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>issues_syntax</th>\n",
       "      <td>-0.434990</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.651852</td>\n",
       "      <td>0.309958</td>\n",
       "      <td>0.379534</td>\n",
       "      <td>0.165053</td>\n",
       "      <td>-0.040641</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.114881</td>\n",
       "      <td>0.390330</td>\n",
       "      <td>...</td>\n",
       "      <td>0.269193</td>\n",
       "      <td>-0.258121</td>\n",
       "      <td>0.378811</td>\n",
       "      <td>0.321158</td>\n",
       "      <td>0.487445</td>\n",
       "      <td>-0.379534</td>\n",
       "      <td>-0.076683</td>\n",
       "      <td>-0.422643</td>\n",
       "      <td>0.202260</td>\n",
       "      <td>-0.354277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>issues_verbs</th>\n",
       "      <td>-0.290389</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.029654</td>\n",
       "      <td>0.100874</td>\n",
       "      <td>0.339471</td>\n",
       "      <td>0.636229</td>\n",
       "      <td>0.156313</td>\n",
       "      <td>0.114881</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.316706</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.193988</td>\n",
       "      <td>-0.241864</td>\n",
       "      <td>-0.047244</td>\n",
       "      <td>-0.356551</td>\n",
       "      <td>-0.065056</td>\n",
       "      <td>-0.445686</td>\n",
       "      <td>-0.073811</td>\n",
       "      <td>-0.010916</td>\n",
       "      <td>-0.166154</td>\n",
       "      <td>-0.009057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>issues_punct</th>\n",
       "      <td>-0.107789</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.300198</td>\n",
       "      <td>0.363085</td>\n",
       "      <td>0.043572</td>\n",
       "      <td>0.102633</td>\n",
       "      <td>0.386810</td>\n",
       "      <td>0.390330</td>\n",
       "      <td>0.316706</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.264654</td>\n",
       "      <td>-0.093981</td>\n",
       "      <td>0.522412</td>\n",
       "      <td>0.193805</td>\n",
       "      <td>0.436126</td>\n",
       "      <td>-0.186938</td>\n",
       "      <td>-0.215475</td>\n",
       "      <td>0.165028</td>\n",
       "      <td>-0.115722</td>\n",
       "      <td>0.133449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>issues_cultur</th>\n",
       "      <td>0.304214</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.152420</td>\n",
       "      <td>0.370345</td>\n",
       "      <td>-0.172929</td>\n",
       "      <td>-0.079760</td>\n",
       "      <td>0.274356</td>\n",
       "      <td>-0.158527</td>\n",
       "      <td>0.102552</td>\n",
       "      <td>0.002768</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.370098</td>\n",
       "      <td>0.087910</td>\n",
       "      <td>-0.273811</td>\n",
       "      <td>-0.218596</td>\n",
       "      <td>-0.465854</td>\n",
       "      <td>-0.270988</td>\n",
       "      <td>-0.517632</td>\n",
       "      <td>0.068317</td>\n",
       "      <td>-0.220860</td>\n",
       "      <td>0.324692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>issues_other</th>\n",
       "      <td>-0.236134</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.398608</td>\n",
       "      <td>-0.178434</td>\n",
       "      <td>0.346438</td>\n",
       "      <td>0.135676</td>\n",
       "      <td>0.589109</td>\n",
       "      <td>-0.435093</td>\n",
       "      <td>0.154625</td>\n",
       "      <td>0.215203</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.124259</td>\n",
       "      <td>-0.331474</td>\n",
       "      <td>-0.018506</td>\n",
       "      <td>-0.007619</td>\n",
       "      <td>-0.204666</td>\n",
       "      <td>-0.209409</td>\n",
       "      <td>-0.334207</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.141771</td>\n",
       "      <td>-0.093912</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>adequacy</th>\n",
       "      <td>0.660220</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.167656</td>\n",
       "      <td>0.190104</td>\n",
       "      <td>-0.379093</td>\n",
       "      <td>-0.077205</td>\n",
       "      <td>0.185563</td>\n",
       "      <td>-0.421055</td>\n",
       "      <td>-0.015792</td>\n",
       "      <td>0.021316</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.484931</td>\n",
       "      <td>0.500959</td>\n",
       "      <td>-0.312355</td>\n",
       "      <td>-0.159545</td>\n",
       "      <td>-0.336249</td>\n",
       "      <td>0.198512</td>\n",
       "      <td>-0.248052</td>\n",
       "      <td>0.386198</td>\n",
       "      <td>-0.404155</td>\n",
       "      <td>0.655534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fluency</th>\n",
       "      <td>0.611124</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.184761</td>\n",
       "      <td>0.244416</td>\n",
       "      <td>-0.423890</td>\n",
       "      <td>-0.009668</td>\n",
       "      <td>-0.320471</td>\n",
       "      <td>-0.252986</td>\n",
       "      <td>-0.107323</td>\n",
       "      <td>-0.321043</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.228843</td>\n",
       "      <td>0.481602</td>\n",
       "      <td>-0.166558</td>\n",
       "      <td>-0.189763</td>\n",
       "      <td>-0.236946</td>\n",
       "      <td>0.370162</td>\n",
       "      <td>-0.067361</td>\n",
       "      <td>0.381757</td>\n",
       "      <td>-0.091372</td>\n",
       "      <td>0.455600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mt_rank_speed_ht</th>\n",
       "      <td>-0.340915</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.132865</td>\n",
       "      <td>0.100437</td>\n",
       "      <td>0.450564</td>\n",
       "      <td>0.139054</td>\n",
       "      <td>-0.104507</td>\n",
       "      <td>0.269193</td>\n",
       "      <td>-0.193988</td>\n",
       "      <td>-0.264654</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.117461</td>\n",
       "      <td>0.281041</td>\n",
       "      <td>0.523723</td>\n",
       "      <td>0.339336</td>\n",
       "      <td>0.218368</td>\n",
       "      <td>0.602091</td>\n",
       "      <td>-0.232173</td>\n",
       "      <td>0.155075</td>\n",
       "      <td>-0.805696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mt_rank_speed_pe</th>\n",
       "      <td>0.741651</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.363105</td>\n",
       "      <td>-0.088226</td>\n",
       "      <td>-0.442500</td>\n",
       "      <td>-0.130292</td>\n",
       "      <td>0.124597</td>\n",
       "      <td>-0.258121</td>\n",
       "      <td>-0.241864</td>\n",
       "      <td>-0.093981</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.117461</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.045920</td>\n",
       "      <td>0.233252</td>\n",
       "      <td>0.119832</td>\n",
       "      <td>0.322521</td>\n",
       "      <td>0.343788</td>\n",
       "      <td>0.714102</td>\n",
       "      <td>-0.117679</td>\n",
       "      <td>0.682867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mt_rank_cog_effort_ht</th>\n",
       "      <td>0.047291</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.077162</td>\n",
       "      <td>0.174987</td>\n",
       "      <td>-0.015152</td>\n",
       "      <td>0.050473</td>\n",
       "      <td>0.382726</td>\n",
       "      <td>0.378811</td>\n",
       "      <td>-0.047244</td>\n",
       "      <td>0.522412</td>\n",
       "      <td>...</td>\n",
       "      <td>0.281041</td>\n",
       "      <td>0.045920</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.343438</td>\n",
       "      <td>0.757108</td>\n",
       "      <td>0.276604</td>\n",
       "      <td>0.232848</td>\n",
       "      <td>0.395591</td>\n",
       "      <td>0.236083</td>\n",
       "      <td>-0.179355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mt_rank_cog_effort_pe</th>\n",
       "      <td>-0.091132</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.040921</td>\n",
       "      <td>0.255202</td>\n",
       "      <td>0.344124</td>\n",
       "      <td>-0.032121</td>\n",
       "      <td>0.172409</td>\n",
       "      <td>0.321158</td>\n",
       "      <td>-0.356551</td>\n",
       "      <td>0.193805</td>\n",
       "      <td>...</td>\n",
       "      <td>0.523723</td>\n",
       "      <td>0.233252</td>\n",
       "      <td>0.343438</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.271134</td>\n",
       "      <td>0.370625</td>\n",
       "      <td>0.171679</td>\n",
       "      <td>0.015762</td>\n",
       "      <td>0.002661</td>\n",
       "      <td>-0.246144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mt_rank_tech_effort_ht</th>\n",
       "      <td>-0.083056</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.322302</td>\n",
       "      <td>-0.026104</td>\n",
       "      <td>0.091474</td>\n",
       "      <td>-0.195161</td>\n",
       "      <td>0.301709</td>\n",
       "      <td>0.487445</td>\n",
       "      <td>-0.065056</td>\n",
       "      <td>0.436126</td>\n",
       "      <td>...</td>\n",
       "      <td>0.339336</td>\n",
       "      <td>0.119832</td>\n",
       "      <td>0.757108</td>\n",
       "      <td>0.271134</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.041378</td>\n",
       "      <td>0.501173</td>\n",
       "      <td>0.279674</td>\n",
       "      <td>0.060484</td>\n",
       "      <td>-0.178153</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mt_rank_tech_effort_pe</th>\n",
       "      <td>0.424611</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.353757</td>\n",
       "      <td>0.025070</td>\n",
       "      <td>-0.365617</td>\n",
       "      <td>-0.219827</td>\n",
       "      <td>-0.163297</td>\n",
       "      <td>-0.379534</td>\n",
       "      <td>-0.445686</td>\n",
       "      <td>-0.186938</td>\n",
       "      <td>...</td>\n",
       "      <td>0.218368</td>\n",
       "      <td>0.322521</td>\n",
       "      <td>0.276604</td>\n",
       "      <td>0.370625</td>\n",
       "      <td>0.041378</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.270421</td>\n",
       "      <td>0.423535</td>\n",
       "      <td>-0.052110</td>\n",
       "      <td>0.031731</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mt_rank_pref_ht</th>\n",
       "      <td>0.067850</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.206631</td>\n",
       "      <td>0.007033</td>\n",
       "      <td>-0.017165</td>\n",
       "      <td>-0.199676</td>\n",
       "      <td>-0.076683</td>\n",
       "      <td>-0.073811</td>\n",
       "      <td>-0.215475</td>\n",
       "      <td>...</td>\n",
       "      <td>0.602091</td>\n",
       "      <td>0.343788</td>\n",
       "      <td>0.232848</td>\n",
       "      <td>0.171679</td>\n",
       "      <td>0.501173</td>\n",
       "      <td>0.270421</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.251275</td>\n",
       "      <td>-0.052907</td>\n",
       "      <td>-0.237864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mt_rank_pref_pe</th>\n",
       "      <td>0.759264</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.440385</td>\n",
       "      <td>-0.105126</td>\n",
       "      <td>-0.497214</td>\n",
       "      <td>-0.094606</td>\n",
       "      <td>0.347242</td>\n",
       "      <td>-0.422643</td>\n",
       "      <td>-0.010916</td>\n",
       "      <td>0.165028</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.232173</td>\n",
       "      <td>0.714102</td>\n",
       "      <td>0.395591</td>\n",
       "      <td>0.015762</td>\n",
       "      <td>0.279674</td>\n",
       "      <td>0.423535</td>\n",
       "      <td>0.251275</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.161023</td>\n",
       "      <td>0.596726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>duration_s</th>\n",
       "      <td>-0.237797</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.298776</td>\n",
       "      <td>-0.354283</td>\n",
       "      <td>0.094627</td>\n",
       "      <td>0.136314</td>\n",
       "      <td>-0.016808</td>\n",
       "      <td>0.202260</td>\n",
       "      <td>-0.166154</td>\n",
       "      <td>-0.115722</td>\n",
       "      <td>...</td>\n",
       "      <td>0.155075</td>\n",
       "      <td>-0.117679</td>\n",
       "      <td>0.236083</td>\n",
       "      <td>0.002661</td>\n",
       "      <td>0.060484</td>\n",
       "      <td>-0.052110</td>\n",
       "      <td>-0.052907</td>\n",
       "      <td>-0.161023</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.184269</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mt_rank_speed_diff</th>\n",
       "      <td>0.693153</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.320488</td>\n",
       "      <td>-0.127190</td>\n",
       "      <td>-0.595383</td>\n",
       "      <td>-0.181125</td>\n",
       "      <td>0.151196</td>\n",
       "      <td>-0.354277</td>\n",
       "      <td>-0.009057</td>\n",
       "      <td>0.133449</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.805696</td>\n",
       "      <td>0.682867</td>\n",
       "      <td>-0.179355</td>\n",
       "      <td>-0.246144</td>\n",
       "      <td>-0.178153</td>\n",
       "      <td>0.031731</td>\n",
       "      <td>-0.237864</td>\n",
       "      <td>0.596726</td>\n",
       "      <td>-0.184269</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>24 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                        mt_rating  diff_mt  issues_order  issues_morph  \\\n",
       "mt_rating                1.000000      NaN     -0.192677      0.131085   \n",
       "diff_mt                       NaN      NaN           NaN           NaN   \n",
       "issues_order            -0.192677      NaN      1.000000      0.419961   \n",
       "issues_morph             0.131085      NaN      0.419961      1.000000   \n",
       "issues_mistrans         -0.597348      NaN      0.265317      0.175491   \n",
       "issues_add              -0.254082      NaN     -0.098844      0.186797   \n",
       "issues_term             -0.024940      NaN     -0.331890     -0.020907   \n",
       "issues_syntax           -0.434990      NaN      0.651852      0.309958   \n",
       "issues_verbs            -0.290389      NaN      0.029654      0.100874   \n",
       "issues_punct            -0.107789      NaN      0.300198      0.363085   \n",
       "issues_cultur            0.304214      NaN     -0.152420      0.370345   \n",
       "issues_other            -0.236134      NaN     -0.398608     -0.178434   \n",
       "adequacy                 0.660220      NaN     -0.167656      0.190104   \n",
       "fluency                  0.611124      NaN     -0.184761      0.244416   \n",
       "mt_rank_speed_ht        -0.340915      NaN      0.132865      0.100437   \n",
       "mt_rank_speed_pe         0.741651      NaN     -0.363105     -0.088226   \n",
       "mt_rank_cog_effort_ht    0.047291      NaN      0.077162      0.174987   \n",
       "mt_rank_cog_effort_pe   -0.091132      NaN     -0.040921      0.255202   \n",
       "mt_rank_tech_effort_ht  -0.083056      NaN      0.322302     -0.026104   \n",
       "mt_rank_tech_effort_pe   0.424611      NaN     -0.353757      0.025070   \n",
       "mt_rank_pref_ht          0.067850      NaN      0.000000     -0.206631   \n",
       "mt_rank_pref_pe          0.759264      NaN     -0.440385     -0.105126   \n",
       "duration_s              -0.237797      NaN     -0.298776     -0.354283   \n",
       "mt_rank_speed_diff       0.693153      NaN     -0.320488     -0.127190   \n",
       "\n",
       "                        issues_mistrans  issues_add  issues_term  \\\n",
       "mt_rating                     -0.597348   -0.254082    -0.024940   \n",
       "diff_mt                             NaN         NaN          NaN   \n",
       "issues_order                   0.265317   -0.098844    -0.331890   \n",
       "issues_morph                   0.175491    0.186797    -0.020907   \n",
       "issues_mistrans                1.000000    0.219827     0.131844   \n",
       "issues_add                     0.219827    1.000000     0.042454   \n",
       "issues_term                    0.131844    0.042454     1.000000   \n",
       "issues_syntax                  0.379534    0.165053    -0.040641   \n",
       "issues_verbs                   0.339471    0.636229     0.156313   \n",
       "issues_punct                   0.043572    0.102633     0.386810   \n",
       "issues_cultur                 -0.172929   -0.079760     0.274356   \n",
       "issues_other                   0.346438    0.135676     0.589109   \n",
       "adequacy                      -0.379093   -0.077205     0.185563   \n",
       "fluency                       -0.423890   -0.009668    -0.320471   \n",
       "mt_rank_speed_ht               0.450564    0.139054    -0.104507   \n",
       "mt_rank_speed_pe              -0.442500   -0.130292     0.124597   \n",
       "mt_rank_cog_effort_ht         -0.015152    0.050473     0.382726   \n",
       "mt_rank_cog_effort_pe          0.344124   -0.032121     0.172409   \n",
       "mt_rank_tech_effort_ht         0.091474   -0.195161     0.301709   \n",
       "mt_rank_tech_effort_pe        -0.365617   -0.219827    -0.163297   \n",
       "mt_rank_pref_ht                0.007033   -0.017165    -0.199676   \n",
       "mt_rank_pref_pe               -0.497214   -0.094606     0.347242   \n",
       "duration_s                     0.094627    0.136314    -0.016808   \n",
       "mt_rank_speed_diff            -0.595383   -0.181125     0.151196   \n",
       "\n",
       "                        issues_syntax  issues_verbs  issues_punct  ...  \\\n",
       "mt_rating                   -0.434990     -0.290389     -0.107789  ...   \n",
       "diff_mt                           NaN           NaN           NaN  ...   \n",
       "issues_order                 0.651852      0.029654      0.300198  ...   \n",
       "issues_morph                 0.309958      0.100874      0.363085  ...   \n",
       "issues_mistrans              0.379534      0.339471      0.043572  ...   \n",
       "issues_add                   0.165053      0.636229      0.102633  ...   \n",
       "issues_term                 -0.040641      0.156313      0.386810  ...   \n",
       "issues_syntax                1.000000      0.114881      0.390330  ...   \n",
       "issues_verbs                 0.114881      1.000000      0.316706  ...   \n",
       "issues_punct                 0.390330      0.316706      1.000000  ...   \n",
       "issues_cultur               -0.158527      0.102552      0.002768  ...   \n",
       "issues_other                -0.435093      0.154625      0.215203  ...   \n",
       "adequacy                    -0.421055     -0.015792      0.021316  ...   \n",
       "fluency                     -0.252986     -0.107323     -0.321043  ...   \n",
       "mt_rank_speed_ht             0.269193     -0.193988     -0.264654  ...   \n",
       "mt_rank_speed_pe            -0.258121     -0.241864     -0.093981  ...   \n",
       "mt_rank_cog_effort_ht        0.378811     -0.047244      0.522412  ...   \n",
       "mt_rank_cog_effort_pe        0.321158     -0.356551      0.193805  ...   \n",
       "mt_rank_tech_effort_ht       0.487445     -0.065056      0.436126  ...   \n",
       "mt_rank_tech_effort_pe      -0.379534     -0.445686     -0.186938  ...   \n",
       "mt_rank_pref_ht             -0.076683     -0.073811     -0.215475  ...   \n",
       "mt_rank_pref_pe             -0.422643     -0.010916      0.165028  ...   \n",
       "duration_s                   0.202260     -0.166154     -0.115722  ...   \n",
       "mt_rank_speed_diff          -0.354277     -0.009057      0.133449  ...   \n",
       "\n",
       "                        mt_rank_speed_ht  mt_rank_speed_pe  \\\n",
       "mt_rating                      -0.340915          0.741651   \n",
       "diff_mt                              NaN               NaN   \n",
       "issues_order                    0.132865         -0.363105   \n",
       "issues_morph                    0.100437         -0.088226   \n",
       "issues_mistrans                 0.450564         -0.442500   \n",
       "issues_add                      0.139054         -0.130292   \n",
       "issues_term                    -0.104507          0.124597   \n",
       "issues_syntax                   0.269193         -0.258121   \n",
       "issues_verbs                   -0.193988         -0.241864   \n",
       "issues_punct                   -0.264654         -0.093981   \n",
       "issues_cultur                  -0.370098          0.087910   \n",
       "issues_other                   -0.124259         -0.331474   \n",
       "adequacy                       -0.484931          0.500959   \n",
       "fluency                        -0.228843          0.481602   \n",
       "mt_rank_speed_ht                1.000000         -0.117461   \n",
       "mt_rank_speed_pe               -0.117461          1.000000   \n",
       "mt_rank_cog_effort_ht           0.281041          0.045920   \n",
       "mt_rank_cog_effort_pe           0.523723          0.233252   \n",
       "mt_rank_tech_effort_ht          0.339336          0.119832   \n",
       "mt_rank_tech_effort_pe          0.218368          0.322521   \n",
       "mt_rank_pref_ht                 0.602091          0.343788   \n",
       "mt_rank_pref_pe                -0.232173          0.714102   \n",
       "duration_s                      0.155075         -0.117679   \n",
       "mt_rank_speed_diff             -0.805696          0.682867   \n",
       "\n",
       "                        mt_rank_cog_effort_ht  mt_rank_cog_effort_pe  \\\n",
       "mt_rating                            0.047291              -0.091132   \n",
       "diff_mt                                   NaN                    NaN   \n",
       "issues_order                         0.077162              -0.040921   \n",
       "issues_morph                         0.174987               0.255202   \n",
       "issues_mistrans                     -0.015152               0.344124   \n",
       "issues_add                           0.050473              -0.032121   \n",
       "issues_term                          0.382726               0.172409   \n",
       "issues_syntax                        0.378811               0.321158   \n",
       "issues_verbs                        -0.047244              -0.356551   \n",
       "issues_punct                         0.522412               0.193805   \n",
       "issues_cultur                       -0.273811              -0.218596   \n",
       "issues_other                        -0.018506              -0.007619   \n",
       "adequacy                            -0.312355              -0.159545   \n",
       "fluency                             -0.166558              -0.189763   \n",
       "mt_rank_speed_ht                     0.281041               0.523723   \n",
       "mt_rank_speed_pe                     0.045920               0.233252   \n",
       "mt_rank_cog_effort_ht                1.000000               0.343438   \n",
       "mt_rank_cog_effort_pe                0.343438               1.000000   \n",
       "mt_rank_tech_effort_ht               0.757108               0.271134   \n",
       "mt_rank_tech_effort_pe               0.276604               0.370625   \n",
       "mt_rank_pref_ht                      0.232848               0.171679   \n",
       "mt_rank_pref_pe                      0.395591               0.015762   \n",
       "duration_s                           0.236083               0.002661   \n",
       "mt_rank_speed_diff                  -0.179355              -0.246144   \n",
       "\n",
       "                        mt_rank_tech_effort_ht  mt_rank_tech_effort_pe  \\\n",
       "mt_rating                            -0.083056                0.424611   \n",
       "diff_mt                                    NaN                     NaN   \n",
       "issues_order                          0.322302               -0.353757   \n",
       "issues_morph                         -0.026104                0.025070   \n",
       "issues_mistrans                       0.091474               -0.365617   \n",
       "issues_add                           -0.195161               -0.219827   \n",
       "issues_term                           0.301709               -0.163297   \n",
       "issues_syntax                         0.487445               -0.379534   \n",
       "issues_verbs                         -0.065056               -0.445686   \n",
       "issues_punct                          0.436126               -0.186938   \n",
       "issues_cultur                        -0.465854               -0.270988   \n",
       "issues_other                         -0.204666               -0.209409   \n",
       "adequacy                             -0.336249                0.198512   \n",
       "fluency                              -0.236946                0.370162   \n",
       "mt_rank_speed_ht                      0.339336                0.218368   \n",
       "mt_rank_speed_pe                      0.119832                0.322521   \n",
       "mt_rank_cog_effort_ht                 0.757108                0.276604   \n",
       "mt_rank_cog_effort_pe                 0.271134                0.370625   \n",
       "mt_rank_tech_effort_ht                1.000000                0.041378   \n",
       "mt_rank_tech_effort_pe                0.041378                1.000000   \n",
       "mt_rank_pref_ht                       0.501173                0.270421   \n",
       "mt_rank_pref_pe                       0.279674                0.423535   \n",
       "duration_s                            0.060484               -0.052110   \n",
       "mt_rank_speed_diff                   -0.178153                0.031731   \n",
       "\n",
       "                        mt_rank_pref_ht  mt_rank_pref_pe  duration_s  \\\n",
       "mt_rating                      0.067850         0.759264   -0.237797   \n",
       "diff_mt                             NaN              NaN         NaN   \n",
       "issues_order                   0.000000        -0.440385   -0.298776   \n",
       "issues_morph                  -0.206631        -0.105126   -0.354283   \n",
       "issues_mistrans                0.007033        -0.497214    0.094627   \n",
       "issues_add                    -0.017165        -0.094606    0.136314   \n",
       "issues_term                   -0.199676         0.347242   -0.016808   \n",
       "issues_syntax                 -0.076683        -0.422643    0.202260   \n",
       "issues_verbs                  -0.073811        -0.010916   -0.166154   \n",
       "issues_punct                  -0.215475         0.165028   -0.115722   \n",
       "issues_cultur                 -0.517632         0.068317   -0.220860   \n",
       "issues_other                  -0.334207         0.000000    0.141771   \n",
       "adequacy                      -0.248052         0.386198   -0.404155   \n",
       "fluency                       -0.067361         0.381757   -0.091372   \n",
       "mt_rank_speed_ht               0.602091        -0.232173    0.155075   \n",
       "mt_rank_speed_pe               0.343788         0.714102   -0.117679   \n",
       "mt_rank_cog_effort_ht          0.232848         0.395591    0.236083   \n",
       "mt_rank_cog_effort_pe          0.171679         0.015762    0.002661   \n",
       "mt_rank_tech_effort_ht         0.501173         0.279674    0.060484   \n",
       "mt_rank_tech_effort_pe         0.270421         0.423535   -0.052110   \n",
       "mt_rank_pref_ht                1.000000         0.251275   -0.052907   \n",
       "mt_rank_pref_pe                0.251275         1.000000   -0.161023   \n",
       "duration_s                    -0.052907        -0.161023    1.000000   \n",
       "mt_rank_speed_diff            -0.237864         0.596726   -0.184269   \n",
       "\n",
       "                        mt_rank_speed_diff  \n",
       "mt_rating                         0.693153  \n",
       "diff_mt                                NaN  \n",
       "issues_order                     -0.320488  \n",
       "issues_morph                     -0.127190  \n",
       "issues_mistrans                  -0.595383  \n",
       "issues_add                       -0.181125  \n",
       "issues_term                       0.151196  \n",
       "issues_syntax                    -0.354277  \n",
       "issues_verbs                     -0.009057  \n",
       "issues_punct                      0.133449  \n",
       "issues_cultur                     0.324692  \n",
       "issues_other                     -0.093912  \n",
       "adequacy                          0.655534  \n",
       "fluency                           0.455600  \n",
       "mt_rank_speed_ht                 -0.805696  \n",
       "mt_rank_speed_pe                  0.682867  \n",
       "mt_rank_cog_effort_ht            -0.179355  \n",
       "mt_rank_cog_effort_pe            -0.246144  \n",
       "mt_rank_tech_effort_ht           -0.178153  \n",
       "mt_rank_tech_effort_pe            0.031731  \n",
       "mt_rank_pref_ht                  -0.237864  \n",
       "mt_rank_pref_pe                   0.596726  \n",
       "duration_s                       -0.184269  \n",
       "mt_rank_speed_diff                1.000000  \n",
       "\n",
       "[24 rows x 24 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "post.corr()"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "f9f85d98b63f393548f3009c8d52d8286e609a1467b1184fe464fb700873fbd3"
  },
  "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": 2
}