From 9e6931cdb628b6f12dd9543027a85144cd25aeba Mon Sep 17 00:00:00 2001
From: Tom Edwards <edwardstj1@cardiff.ac.uk>
Date: Wed, 4 Sep 2024 14:52:07 +0100
Subject: [PATCH] Work towards USAS + ner
---
func/ner/ner.py | 7 +++++++
func/usas/usas.py | 23 ++++++++++++++++++++---
main.py | 10 +++++++---
3 files changed, 34 insertions(+), 6 deletions(-)
diff --git a/func/ner/ner.py b/func/ner/ner.py
index 7777b73..b7a4e51 100644
--- a/func/ner/ner.py
+++ b/func/ner/ner.py
@@ -14,6 +14,13 @@ def run_ner_on_text(page):
word = item.word
ner = item.ner
idx = item.idx
+ print(type(idx))
+ print(idx)
+ print(type(ner))
+ print(ner)
+ print(type(word))
+ print(word)
+ print('--------')
obj = {"word": word, "translation": "", "ner": ner, "idx": idx}
data.append(obj)
diff --git a/func/usas/usas.py b/func/usas/usas.py
index c0ca4fe..455792f 100644
--- a/func/usas/usas.py
+++ b/func/usas/usas.py
@@ -11,10 +11,27 @@ def run_usas_on_text(page):
nlp.add_pipe('pymusas_rule_based_tagger', source=chinese_tagger_pipeline)
output_doc = nlp(page)
+ data = []
- print(f'Text\tPOS\tUSAS Tags')
+ tags = []
+
+ print(f'Text\tPOS\tMWE start and end index\tUSAS Tags')
for token in output_doc:
- print(f'{token.text}\t{token.pos_}\t{token._.pymusas_tags}')
+ start, end = token._.pymusas_mwe_indexes[0]
+ idx = (start, end)
+
+ for el in token._.pymusas_tags:
+ obj = {"word": token.text, "Usas Tags": el, "idx": idx}
+ tags.append(el)
+ data.append(obj)
+
+ res = []
+ procTags = []
+ for x in tags:
+ if x not in procTags:
+ res.append({"Tag": x, "Count": tags.count(x)})
+
+ procTags.append(x)
- result = {'output': "Hello USAS", 'message': 'Done', 'code': 'SUCCESS'}
+ result = {'output': res, 'message': 'Done', 'code': 'SUCCESS'}
return result
diff --git a/main.py b/main.py
index a30a228..34bb595 100644
--- a/main.py
+++ b/main.py
@@ -19,18 +19,22 @@ def test():
return "Success"
-@app.route('/ner', methods=['POST'])
+@app.route("/ner", methods=['POST'])
+
def ner():
request_data = request.get_json()
- print(request_data)
+ #print(request_data)
page = request_data['page']
result = get_ner_for_data(page)
return result
-@app.route("/usas")
+@app.route("/usas", methods=['POST'])
def usas():
+ print('hello')
request_data = request.get_json()
+ print(request_data)
page = request_data['page']
result = get_usas_for_data(page)
+
return result
--
GitLab