Merge branch 'develop' into spacy.io

spacy/util.py

@@ -315,7 +315,7 @@ def read_regex(path):
 
 
 def compile_prefix_regex(entries):
-    """Compile a list of prefix rules into a regex object.
+    """Compile a sequence of prefix rules into a regex object.
 
     entries (tuple): The prefix rules, e.g. spacy.lang.punctuation.TOKENIZER_PREFIXES.
     RETURNS (regex object): The regex object. to be used for Tokenizer.prefix_search.
@@ -332,7 +332,7 @@ def compile_prefix_regex(entries):
 
 
 def compile_suffix_regex(entries):
-    """Compile a list of suffix rules into a regex object.
+    """Compile a sequence of suffix rules into a regex object.
 
     entries (tuple): The suffix rules, e.g. spacy.lang.punctuation.TOKENIZER_SUFFIXES.
     RETURNS (regex object): The regex object. to be used for Tokenizer.suffix_search.
@@ -342,7 +342,7 @@ def compile_suffix_regex(entries):
 
 
 def compile_infix_regex(entries):
-    """Compile a list of infix rules into a regex object.
+    """Compile a sequence of infix rules into a regex object.
 
     entries (tuple): The infix rules, e.g. spacy.lang.punctuation.TOKENIZER_INFIXES.
     RETURNS (regex object): The regex object. to be used for Tokenizer.infix_finditer.

website/docs/usage/processing-pipelines.md

@@ -29,10 +29,10 @@ components. spaCy then does the following:
 >
 > ```json
 > {
->   "name": "example_model",
 >   "lang": "en",
+>   "name": "core_web_sm",
 >   "description": "Example model for spaCy",
->   "pipeline": ["tagger", "parser"]
+>   "pipeline": ["tagger", "parser", "ner"]
 > }
 > ```
 
@@ -51,11 +51,11 @@ components. spaCy then does the following:
 So when you call this...
 
 ```python
-nlp = spacy.load("en")
+nlp = spacy.load("en_core_web_sm")
 ```
 
-... the model tells spaCy to use the language `"en"` and the pipeline
-`["tagger", "parser", "ner"]`. spaCy will then initialize
+... the model's `meta.json` tells spaCy to use the language `"en"` and the
+pipeline `["tagger", "parser", "ner"]`. spaCy will then initialize
 `spacy.lang.en.English`, and create each pipeline component and add it to the
 processing pipeline. It'll then load in the model's data from its data directory
 and return the modified `Language` class for you to use as the `nlp` object.