Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -13,6 +13,8 @@ import pickle
|
|
13 |
from langchain.chat_models import ChatOpenAI
|
14 |
from langchain.prompts.chat import ChatPromptTemplate
|
15 |
|
|
|
|
|
16 |
# Загружаем переменные окружения
|
17 |
load_dotenv()
|
18 |
hf_key = os.getenv("HF_KEY")
|
@@ -38,7 +40,7 @@ def update_faiss_index():
|
|
38 |
]
|
39 |
|
40 |
dataset = concatenate_datasets([train_dataset, test_dataset])
|
41 |
-
dataset = dataset.select(range(5)) # Для тестирования на небольшом количестве данных
|
42 |
docs = []
|
43 |
unique_chunks = set()
|
44 |
for row in tqdm(dataset, desc="Загрузка документов..."):
|
@@ -107,7 +109,7 @@ def retrieve_articles(query, language):
|
|
107 |
if language == "Russian":
|
108 |
translated_query = translate_ru_uz(query)
|
109 |
else:
|
110 |
-
translated_query = query
|
111 |
|
112 |
results = db.similarity_search(translated_query, k=3)
|
113 |
|
|
|
13 |
from langchain.chat_models import ChatOpenAI
|
14 |
from langchain.prompts.chat import ChatPromptTemplate
|
15 |
|
16 |
+
from uz_translit import to_latin
|
17 |
+
|
18 |
# Загружаем переменные окружения
|
19 |
load_dotenv()
|
20 |
hf_key = os.getenv("HF_KEY")
|
|
|
40 |
]
|
41 |
|
42 |
dataset = concatenate_datasets([train_dataset, test_dataset])
|
43 |
+
# dataset = dataset.select(range(5)) # Для тестирования на небольшом количестве данных
|
44 |
docs = []
|
45 |
unique_chunks = set()
|
46 |
for row in tqdm(dataset, desc="Загрузка документов..."):
|
|
|
109 |
if language == "Russian":
|
110 |
translated_query = translate_ru_uz(query)
|
111 |
else:
|
112 |
+
translated_query = to_lati(query)
|
113 |
|
114 |
results = db.similarity_search(translated_query, k=3)
|
115 |
|