Update app.py
Browse files
app.py
CHANGED
|
@@ -1,3 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
from langchain.text_splitter import CharacterTextSplitter
|
| 2 |
from langchain.embeddings import HuggingFaceBgeEmbeddings
|
| 3 |
from langchain.vectorstores import FAISS
|
|
@@ -51,12 +56,13 @@ def translate(userinput, target_lang, source_lang=None):
|
|
| 51 |
translation = result['translated_text']
|
| 52 |
return source_lange, translation
|
| 53 |
|
| 54 |
-
def get_pdf_text(
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
text
|
|
|
|
| 60 |
return text
|
| 61 |
|
| 62 |
|
|
@@ -96,13 +102,11 @@ def get_conversation_chain(vectorstore:FAISS) -> ConversationalRetrievalChain:
|
|
| 96 |
def main():
|
| 97 |
st.title("SemaNaPDF📚")
|
| 98 |
# upload file
|
| 99 |
-
|
| 100 |
-
|
| 101 |
-
)
|
| 102 |
-
if pdf_docs is not None:
|
| 103 |
with st.spinner("processing"):
|
| 104 |
# get pdf text
|
| 105 |
-
raw_text = get_pdf_text(
|
| 106 |
|
| 107 |
# get the text chunks
|
| 108 |
text_chunks = get_text_chunks(raw_text)
|
|
@@ -133,6 +137,16 @@ def main():
|
|
| 133 |
st.markdown(response)
|
| 134 |
st.session_state.messages.append({"role": "assistant", "content": response})
|
| 135 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 136 |
|
| 137 |
if __name__ == '__main__':
|
| 138 |
main()
|
|
|
|
| 1 |
+
"""
|
| 2 |
+
creator: Lewis Kamau Kimaru
|
| 3 |
+
Function: chat with pdf documents in different languages
|
| 4 |
+
|
| 5 |
+
"""
|
| 6 |
from langchain.text_splitter import CharacterTextSplitter
|
| 7 |
from langchain.embeddings import HuggingFaceBgeEmbeddings
|
| 8 |
from langchain.vectorstores import FAISS
|
|
|
|
| 56 |
translation = result['translated_text']
|
| 57 |
return source_lange, translation
|
| 58 |
|
| 59 |
+
def get_pdf_text(pdf : Union[str, bytes, bytearray]) -> str:
|
| 60 |
+
reader = PdfReader(pdf)
|
| 61 |
+
pdf_text = ''
|
| 62 |
+
for page in (reader.pages):
|
| 63 |
+
text = page.extract_text()
|
| 64 |
+
if text:
|
| 65 |
+
pdf_text += text
|
| 66 |
return text
|
| 67 |
|
| 68 |
|
|
|
|
| 102 |
def main():
|
| 103 |
st.title("SemaNaPDF📚")
|
| 104 |
# upload file
|
| 105 |
+
pdf = st.file_uploader("Upload a PDF Document", type="pdf")
|
| 106 |
+
if pdf is not None:
|
|
|
|
|
|
|
| 107 |
with st.spinner("processing"):
|
| 108 |
# get pdf text
|
| 109 |
+
raw_text = get_pdf_text(pdf)
|
| 110 |
|
| 111 |
# get the text chunks
|
| 112 |
text_chunks = get_text_chunks(raw_text)
|
|
|
|
| 137 |
st.markdown(response)
|
| 138 |
st.session_state.messages.append({"role": "assistant", "content": response})
|
| 139 |
|
| 140 |
+
# Signature
|
| 141 |
+
st.markdown(
|
| 142 |
+
"""
|
| 143 |
+
<div style="position: fixed; bottom: 0; right: 0; padding: 10px;">
|
| 144 |
+
<a href="https://kamaukimaru.vercel.app" target="_blank" rel="noopener noreferrer" style="font-size: 12px; color: #808080; text-decoration: none;">©2023 Lewis Kimaru. All rights reserved.</a>
|
| 145 |
+
</div>
|
| 146 |
+
""",
|
| 147 |
+
unsafe_allow_html=True
|
| 148 |
+
)
|
| 149 |
+
|
| 150 |
|
| 151 |
if __name__ == '__main__':
|
| 152 |
main()
|