{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from transformers import pipeline\n",
    "\n",
    "fill_mask = pipeline(\n",
    "    \"fill-mask\",\n",
    "    \"latin_BERT_final\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'score': 0.010115200653672218,\n",
       "  'token': 265,\n",
       "  'token_str': 'hoc',\n",
       "  'sequence': 'roma in hoc est.'},\n",
       " {'score': 0.004335678182542324,\n",
       "  'token': 1156,\n",
       "  'token_str': 'deo',\n",
       "  'sequence': 'roma in deo est.'},\n",
       " {'score': 0.003660168731585145,\n",
       "  'token': 146,\n",
       "  'token_str': 'non',\n",
       "  'sequence': 'roma in non est.'},\n",
       " {'score': 0.0034285704605281353,\n",
       "  'token': 745,\n",
       "  'token_str': 'nobis',\n",
       "  'sequence': 'roma in nobis est.'},\n",
       " {'score': 0.0032894855830818415,\n",
       "  'token': 971,\n",
       "  'token_str': 'rebus',\n",
       "  'sequence': 'roma in rebus est.'}]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fill_mask(\"Roma in [MASK] est.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'score': 0.04990068078041077,\n",
       "  'token': 145,\n",
       "  'token_str': 'est',\n",
       "  'sequence': 'ubi est est?.'},\n",
       " {'score': 0.01739734411239624,\n",
       "  'token': 215,\n",
       "  'token_str': 'quid',\n",
       "  'sequence': 'ubi est quid?.'},\n",
       " {'score': 0.008733403868973255,\n",
       "  'token': 391,\n",
       "  'token_str': 'mihi',\n",
       "  'sequence': 'ubi est mihi?.'},\n",
       " {'score': 0.007146364543586969,\n",
       "  'token': 368,\n",
       "  'token_str': 'sum',\n",
       "  'sequence': 'ubi est sum?.'},\n",
       " {'score': 0.006486538797616959,\n",
       "  'token': 425,\n",
       "  'token_str': 'tibi',\n",
       "  'sequence': 'ubi est tibi?.'}]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fill_mask(\"Ubi est [MASK] ?.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'score': 0.7615262269973755,\n",
       "  'token': 12,\n",
       "  'token_str': '.',\n",
       "  'sequence': 'de honoratorum.'},\n",
       " {'score': 0.03385818004608154,\n",
       "  'token': 23,\n",
       "  'token_str': ':',\n",
       "  'sequence': 'de honoratorum :'},\n",
       " {'score': 0.02129465527832508,\n",
       "  'token': 10,\n",
       "  'token_str': ',',\n",
       "  'sequence': 'de honoratorum,'},\n",
       " {'score': 0.014383483678102493,\n",
       "  'token': 25,\n",
       "  'token_str': '?',\n",
       "  'sequence': 'de honoratorum?'},\n",
       " {'score': 0.008870471268892288,\n",
       "  'token': 109,\n",
       "  'token_str': 'et',\n",
       "  'sequence': 'de honoratorum et'}]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# De honoratorum vehiculis\n",
    "fill_mask(\"De honoratorum [MASK]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'score': 0.00837777741253376,\n",
       "  'token': 1838,\n",
       "  'token_str': 'urbe',\n",
       "  'sequence': 'gallia est omnis divisa in urbe tres'},\n",
       " {'score': 0.007593569345772266,\n",
       "  'token': 1628,\n",
       "  'token_str': 'corpore',\n",
       "  'sequence': 'gallia est omnis divisa in corpore tres'},\n",
       " {'score': 0.007336211856454611,\n",
       "  'token': 2035,\n",
       "  'token_str': 'medio',\n",
       "  'sequence': 'gallia est omnis divisa in medio tres'},\n",
       " {'score': 0.006218622904270887,\n",
       "  'token': 983,\n",
       "  'token_str': 'parte',\n",
       "  'sequence': 'gallia est omnis divisa in parte tres'},\n",
       " {'score': 0.0054352362640202045,\n",
       "  'token': 238,\n",
       "  'token_str': 'quo',\n",
       "  'sequence': 'gallia est omnis divisa in quo tres'}]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fill_mask(\"Gallia est omnis divisa in [MASK] tres\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'score': 0.008044794201850891,\n",
       "  'token': 1628,\n",
       "  'token_str': 'corpore',\n",
       "  'sequence': 'gallia est omnis divisa in corpore tres.'},\n",
       " {'score': 0.00732386251911521,\n",
       "  'token': 1838,\n",
       "  'token_str': 'urbe',\n",
       "  'sequence': 'gallia est omnis divisa in urbe tres.'},\n",
       " {'score': 0.0072334203869104385,\n",
       "  'token': 983,\n",
       "  'token_str': 'parte',\n",
       "  'sequence': 'gallia est omnis divisa in parte tres.'},\n",
       " {'score': 0.006316048558801413,\n",
       "  'token': 2035,\n",
       "  'token_str': 'medio',\n",
       "  'sequence': 'gallia est omnis divisa in medio tres.'},\n",
       " {'score': 0.004988126456737518,\n",
       "  'token': 1177,\n",
       "  'token_str': 'terra',\n",
       "  'sequence': 'gallia est omnis divisa in terra tres.'}]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Adding a point changes the predictions!\n",
    "fill_mask(\"Gallia est omnis divisa in [MASK] tres.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "bertenv",
   "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": "3.10.6"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}