From b801dbf8c01f5537805721fa27b2ec472fe1f58a Mon Sep 17 00:00:00 2001
From: Vincent <vincentwarmerdam@gmail.com>
Date: Thu, 27 Jan 2022 12:20:31 +0100
Subject: [PATCH] fix flashtext docs

---
 docs/docs/extractors/flashtext.md | 36 +++++++++++++++----------------
 1 file changed, 18 insertions(+), 18 deletions(-)

diff --git a/docs/docs/extractors/flashtext.md b/docs/docs/extractors/flashtext.md
index e0e08ea..24f600e 100644
--- a/docs/docs/extractors/flashtext.md
+++ b/docs/docs/extractors/flashtext.md
@@ -9,12 +9,12 @@
     ```
 
 This entity extractor uses the [flashtext](https://flashtext.readthedocs.io/en/latest/) library
-to extract entities using [lookup tables](https://rasa.com/docs/rasa/nlu-training-data#lookup-tables).
+to extract entities.
 
 This is similar to [RegexEntityExtractor](https://rasa.com/docs/rasa/components#regexentityextractor), but
 different in a few ways:
 
-1. `FlashTextEntityExtractor` takes only `lookups`, **not** regex patterns
+1. `FlashTextEntityExtractor` uses token-matching to find entities, **not** regex patterns
 2. `FlashTextEntityExtractor` matches using whitespace word boundaries. You cannot set it
 to match words regardless of boundaries.
 3. `FlashTextEntityExtractor` is *much* faster than `RegexEntityExtractor`. This is especially true
@@ -25,6 +25,8 @@ use the parameter `non_word_boundaries`
 
 ## Configurable Variables
 
+- **path**: the path to the lookup text file
+- **entity_name**: the name of the entity to attach to the message
 - **case_sensitive**: whether to consider case when matching entities. `False` by default.
 - **non_word_boundaries**: characters which shouldn't be considered word boundaries.
 
@@ -42,26 +44,24 @@ pipeline:
   min_ngram: 1
   max_ngram: 4
 - name: rasa_nlu_examples.extractors.FlashTextEntityExtractor
-  case_sensitive: True
-  non_word_boundary:
-  - "_"
-  - ","
+  case_sensitive: False
+  path: path/to/file.txt
+  entity_name: country
 - name: DIETClassifier
   epochs: 100
 ```
-You must include [lookup tables](https://rasa.com/docs/rasa/nlu-training-data#lookup-tables) in your NLU data. This
-might look like:
+You must include a plain text file that contains the tokens to detect.
+Such a file might look like:
+
 ```yaml
-nlu:
-- lookup: country
-  examples: |
-    - Afghanistan
-    - Albania
-    - ...
-    - Zambia
-    - Zimbabwe
+Afghanistan
+Albania
+...
+Zambia
+Zimbabwe
 ```
-In this example, anytime a user's utterance contains an exact match for a country from the lookup table above,
+
+In this example, anytime a user's utterance contains an exact match for a country,
 `FlashTextEntityExtractor` will extract this as an entity with type `country`. You should include a few examples with
 this entity in your intent data, like so:
 
@@ -69,5 +69,5 @@ this entity in your intent data, like so:
 - intent: inform_home_country
   examples: |
     - I am from [Afghanistan](country)
-    - My family is from [Albania](country
+    - My family is from [Albania](country)
 ```