Update and rerun notebooks under tutorials/corpus

CODAIT · Feb 24, 2021 · cbc0d2b · cbc0d2b
1 parent 98dc855
commit cbc0d2b
Show file tree

Hide file tree

Showing 5 changed files with 2,211 additions and 1,640 deletions.
diff --git a/tutorials/corpus/CoNLL_2.ipynb b/tutorials/corpus/CoNLL_2.ipynb
@@ -2543,9 +2543,7 @@
  " pd.DataFrame({\n",
  " \"fold\": k[0], # Keys are (collection, offset) tuples\n",
  " \"doc_offset\": k[1],\n",
- " # TokenSpanArrays from different documents can't currently be stacked,\n",
- " # so convert to TokenSpan objects.\n",
- " \"span\" : indicators[k][\"span\"].astype(object),\n",
+ " \"span\" : indicators[k][\"span\"],\n",
  " \"ent_type\": indicators[k][\"ent_type\"],\n",
  " \"gold\": indicators[k][\"gold\"],\n",
  " \"num_teams\": indicators[k][\"num_teams\"]\n",
@@ -3730,10 +3728,10 @@
  "data": {
  "text/html": [
  "\n",
- " <div id=\"spanArray\">\n",
- "  <div id=\"spans\" \n",
- "  style=\"background-color:#F0F0F0; border: 1px solid #E0E0E0; float:left; padding:10px;\">\n",
- "  <table border=\"1\" class=\"dataframe\">\n",
+ "<div id=\"spanArray\">\n",
+ " <div id=\"spans\" \n",
+ " style=\"background-color:#F0F0F0; border: 1px solid #E0E0E0; float:left; padding:10px;\">\n",
+ " <table border=\"1\" class=\"dataframe\">\n",
  " <thead>\n",
  " <tr style=\"text-align: right;\">\n",
  " <th></th>\n",
@@ -3899,17 +3897,18 @@
  " </tr>\n",
  " </tbody>\n",
  "</table>\n",
- " </div>\n",
- " <div id=\"text\"\n",
- " style=\"float:right; background-color:#F5F5F5; border: 1px solid #E0E0E0; width: 60%;\">\n",
+ " </div>\n",
+ " <div id=\"text\"\n",
+ " style=\"float:right; background-color:#F5F5F5; border: 1px solid #E0E0E0; width: 60%;\">\n",
+ "\n",
  " <div style=\"float:center; padding:10px\">\n",
  " <p style=\"font-family:monospace\">\n",
  " -DOCSTART-<br><span style=\"background-color:yellow\">Belgian</span> police smash major drugs rings, 30 arrested.<br><span style=\"background-color:yellow\">BRUSSELS</span> 1996-12-06<br>Police smashed two drugs smuggling rings and arrested 30 people after a taxidriver in <span style=\"background-color:yellow\">Spain</span> alerted them to a suitcase of heroin left in his cab, <span style=\"background-color:yellow\">Belgian</span> police said on Friday.<br>Police seized dozens of kilos of heroin with a street value of hundreds of millions of <span style=\"background-color:yellow\">Belgian</span> francs, a public prosecutor&#39;s office spokesman in the port city of <span style=\"background-color:yellow\">Antwerp</span> said.<br>He said a 24-year-old <span style=\"background-color:yellow\">Belgian</span> woman left a suitcase containing 13 kg (29 lb) of heroin in a taxi in <span style=\"background-color:yellow\">Barcelona</span>.<br>The taxidriver alerted police who arrested a 33-year-old <span style=\"background-color:yellow\">Turkish</span> man when he came to pick up the suitcase at a lost luggage office.<br>The woman was later arrested in <span style=\"background-color:yellow\">Belgium</span>.<br>She and the <span style=\"background-color:yellow\">Turkish</span> man smuggled heroin from <span style=\"background-color:yellow\">Turkey</span> to <span style=\"background-color:yellow\">Antwerp</span> from where it was taken to <span style=\"background-color:yellow\">Spain</span>, <span style=\"background-color:yellow\">France</span> and <span style=\"background-color:yellow\">Germany</span> by others, the spokesman said.<br>He said 14 people were arrested in <span style=\"background-color:yellow\">Belgium</span> and 16 others in other <span style=\"background-color:yellow\">European</span> nations after an investigation lasting nearly a year.<br>(<span>&#36;</span>1=32.14 <span style=\"background-color:yellow\">Belgian</span> Franc)\n",
  " </p>\n",
  " </div>\n",
- " </div>\n",
- "  </div>\n",
- " "
+ "\n",
+ " </div>\n",
+ "</div>\n"
  ],
  "text/plain": [
  "<TokenSpanArray>\n",
@@ -3986,9 +3985,7 @@
  " pd.DataFrame({\n",
  " \"fold\": k[0], # Keys are (collection, offset) tuples\n",
  " \"doc_offset\": k[1],\n",
- " # TokenSpanArrays from different documents can't currently be stacked,\n",
- " # so convert to TokenSpan objects.\n",
- " \"span\" : dev_indicators[k][\"span\"].astype(object),\n",
+ " \"span\" : dev_indicators[k][\"span\"],\n",
  " \"ent_type\": dev_indicators[k][\"ent_type\"],\n",
  " \"gold\": dev_indicators[k][\"gold\"],\n",
  " \"num_teams\": dev_indicators[k][\"num_teams\"]\n",