From a7aacfa323cd5596fe0a590962d2e0e5d35c386c Mon Sep 17 00:00:00 2001
From: maksim <kashin20031984@mail.ru>
Date: Tue, 4 Jun 2024 00:17:40 +0400
Subject: [PATCH] =?UTF-8?q?=D0=94=D0=BE=D0=B1=D0=B0=D0=B2=D0=BB=D0=B8=20?=
 =?UTF-8?q?=D1=88=D1=82=D1=83=D0=BA=D1=83,=20=D0=BA=D0=BE=D1=82=D0=BE?=
 =?UTF-8?q?=D1=80=D0=B0=D1=8F=20=D0=BF=D1=80=D0=BE=D1=81=D1=82=D0=BE=20?=
 =?UTF-8?q?=D1=81=D0=BE=D1=80=D1=82=D0=B8=D1=80=D1=83=D0=B5=D1=82.=20?=
 =?UTF-8?q?=D0=92=D1=81=D0=B5=20=D1=80=D0=B0=D0=B1=D0=BE=D1=82=D1=82=D0=B0?=
 =?UTF-8?q?=D0=B5=D1=82,=20=D0=BE=D1=81=D1=82=D0=BB=D0=BE=D1=81=D1=8C=20?=
 =?UTF-8?q?=D0=BD=D0=BE=D1=80=D0=BC=20=D0=B2=20=D0=BF=D1=80=D0=BE=D0=B5?=
 =?UTF-8?q?=D0=BA=D1=82=20=D1=81=D1=83=D0=BD=D1=83=D1=82=D1=8C=20=D0=B8=20?=
 =?UTF-8?q?=D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0=D1=82=D1=8C=20API?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 neural_network/dataset/sortTest.py | 51 ++++++++++++++++++++++++++++++
 1 file changed, 51 insertions(+)
 create mode 100644 neural_network/dataset/sortTest.py

diff --git a/neural_network/dataset/sortTest.py b/neural_network/dataset/sortTest.py
new file mode 100644
index 0000000..d312471
--- /dev/null
+++ b/neural_network/dataset/sortTest.py
@@ -0,0 +1,51 @@
+import json
+import pandas as pd
+
+# JSON данные
+json_data = '''
+[
+  { "value": "AAQ", "label": "Анапа, Витязево, AAQ", "timezone": "+3", "latitude": 45.0029, "longitude": 37.3473 },
+  { "value": "ARH", "label": "Архангельск, Талаги, ARH", "timezone": "+3", "latitude": 64.6003, "longitude": 40.7168 },
+  { "value": "ASF", "label": "Астрахань, им. Б.Н. Кустодиева, ASF", "timezone": "+4", "latitude": 46.2833, "longitude": 48.0063 },
+  { "value": "BAX", "label": "Барнаул, Михайловка, BAX", "timezone": "+7", "latitude": 53.3638, "longitude": 83.5385 }
+]
+'''
+
+data = json.loads(json_data)
+
+# Извлекаем первое слово перед запятой и первое слово после запятой
+cities_set = set()
+for entry in data:
+    parts = entry['label'].split(',')
+    if len(parts) > 1:
+        city1 = parts[0].strip().split()[0] if parts[0].strip().split() else ''
+        city2 = parts[1].strip().split()[0] if parts[1].strip().split() else ''
+        cities_set.update([city1, city2])
+
+# Прочитаем CSV файл
+csv_file_path = 'geo-reviews-dataset-2023.csv'
+df = pd.read_csv(csv_file_path)
+
+# Функция для проверки совпадения города в адресе
+def find_city(address, cities_set):
+    parts = address.split(',')
+    if len(parts) > 1:
+        word1 = parts[0].strip().split()[0] if parts[0].strip().split() else ''
+        word2 = parts[1].strip().split()[0] if parts[1].strip().split() else ''
+        if word1 in cities_set:
+            return word1
+        elif word2 in cities_set:
+            return word2
+    return None
+
+# Добавим новый столбец на основе первого слова из адреса
+df['city'] = df['address'].apply(lambda x: find_city(x, cities_set))
+
+# Оставим только те строки, где город из CSV файла совпадает с городом из JSON
+df_filtered = df[df['city'].notnull()]
+
+# Сохраним отсортированный DataFrame обратно в CSV
+output_file_path = 'sorted_filtered_geo-reviews-dataset-2023.csv'
+df_filtered.to_csv(output_file_path, index=False)
+
+print(df_filtered[:15])