Merge branch 'develop'

examples/text/CoNLL2002_Dutch-BiLSTM.ipynb

@@ -169,6 +169,13 @@
     "learner.plot('lr')"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "As shown below, our model achieves an F1-Sccore of 74.30. While this is lower than our NER results for other languages like [English, Chinese, and Russian](https://nbviewer.jupyter.org/github/amaiya/ktrain/blob/master/examples/text/shallownlp-examples.ipynb), it appears to be higher than the [Spacy Dutch NER model](https://spacy.io/models/nl)."
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 14,
@@ -257,6 +264,13 @@
     "predictor.save('/tmp/my_dutch_nermodel')"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "As of v0.11.0 of *ktrain*, saved NER predictors can also be used with the [shallownlp NER](https://nbviewer.jupyter.org/github/amaiya/ktrain/blob/master/examples/text/shallownlp-examples.ipynb) class to make predictions on new sentences."
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 28,
@@ -283,7 +297,7 @@
     }
    ],
    "source": [
-    "ner.predict(dutch_text)"
+    "ner.predict(dutch_text, merge_tokens=True)"
    ]
   }
  ],