Spaces:
Build error
Build error
Update app.py
Browse files
app.py
CHANGED
@@ -1,3 +1,4 @@
|
|
|
|
1 |
import os
|
2 |
import re
|
3 |
import numpy as np
|
@@ -12,12 +13,7 @@ from typing import List, Dict, Any, Tuple, Optional, Union, Generator
|
|
12 |
from dataclasses import dataclass
|
13 |
import gradio as gr
|
14 |
|
15 |
-
#
|
16 |
-
os.system("pip install -q pymupdf langchain langchain_community sentence-transformers faiss-cpu huggingface_hub")
|
17 |
-
os.system("pip install -q llama-cpp-python transformers rank_bm25 nltk")
|
18 |
-
os.system("pip install -q git+https://github.com/chroma-core/chroma.git")
|
19 |
-
|
20 |
-
# Import dependencies after installation
|
21 |
import fitz # PyMuPDF
|
22 |
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
23 |
from langchain_community.vectorstores import FAISS
|
@@ -29,6 +25,13 @@ from nltk.tokenize import word_tokenize
|
|
29 |
from nltk.corpus import stopwords
|
30 |
from huggingface_hub import hf_hub_download
|
31 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
# Setup directories for Spaces
|
33 |
os.makedirs("pdfs", exist_ok=True)
|
34 |
os.makedirs("models", exist_ok=True)
|
|
|
1 |
+
|
2 |
import os
|
3 |
import re
|
4 |
import numpy as np
|
|
|
13 |
from dataclasses import dataclass
|
14 |
import gradio as gr
|
15 |
|
16 |
+
# Import dependencies (no need for pip install commands)
|
|
|
|
|
|
|
|
|
|
|
17 |
import fitz # PyMuPDF
|
18 |
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
19 |
from langchain_community.vectorstores import FAISS
|
|
|
25 |
from nltk.corpus import stopwords
|
26 |
from huggingface_hub import hf_hub_download
|
27 |
|
28 |
+
# Download nltk resources
|
29 |
+
try:
|
30 |
+
nltk.download('punkt', quiet=True)
|
31 |
+
nltk.download('stopwords', quiet=True)
|
32 |
+
except:
|
33 |
+
print("Failed to download NLTK resources, continuing without them")
|
34 |
+
|
35 |
# Setup directories for Spaces
|
36 |
os.makedirs("pdfs", exist_ok=True)
|
37 |
os.makedirs("models", exist_ok=True)
|