Spaces:
Sleeping
Sleeping
Commit
•
1c52547
1
Parent(s):
5a2a128
adding online PDF loader
Browse files
app.py
CHANGED
@@ -48,13 +48,13 @@ def document_loader(file_data,api_key,doc_type='pdf',llm='Huggingface'):
|
|
48 |
embedding_model = SentenceTransformerEmbeddings(model_name='all-mpnet-base-v2',model_kwargs={"device": DEVICE})
|
49 |
document = None
|
50 |
if doc_type == 'pdf':
|
51 |
-
document = process_pdf_document(
|
52 |
elif doc_type == 'text':
|
53 |
-
document = process_text_document(
|
54 |
elif doc_type == 'csv':
|
55 |
-
document = process_csv_document(
|
56 |
elif doc_type == 'word':
|
57 |
-
document = process_word_document(
|
58 |
if document:
|
59 |
texts = process_documents(documents=document)
|
60 |
vector_db = FAISS.from_documents(documents=texts, embedding= embedding_model)
|
@@ -71,20 +71,20 @@ def document_loader(file_data,api_key,doc_type='pdf',llm='Huggingface'):
|
|
71 |
return "Ready..."
|
72 |
|
73 |
|
74 |
-
def process_text_document(
|
75 |
-
loader = TextLoader(
|
76 |
document = loader.load()
|
77 |
return document
|
78 |
|
79 |
|
80 |
-
def process_csv_document(
|
81 |
-
loader = CSVLoader(file_path=
|
82 |
document = loader.load()
|
83 |
return document
|
84 |
|
85 |
|
86 |
-
def process_word_document(
|
87 |
-
loader = UnstructuredWordDocumentLoader(file_path=
|
88 |
document = loader.load()
|
89 |
return document
|
90 |
|
|
|
48 |
embedding_model = SentenceTransformerEmbeddings(model_name='all-mpnet-base-v2',model_kwargs={"device": DEVICE})
|
49 |
document = None
|
50 |
if doc_type == 'pdf':
|
51 |
+
document = process_pdf_document(document_file=file_data)
|
52 |
elif doc_type == 'text':
|
53 |
+
document = process_text_document(document_file=file_data)
|
54 |
elif doc_type == 'csv':
|
55 |
+
document = process_csv_document(document_file=file_data)
|
56 |
elif doc_type == 'word':
|
57 |
+
document = process_word_document(document_file=file_data)
|
58 |
if document:
|
59 |
texts = process_documents(documents=document)
|
60 |
vector_db = FAISS.from_documents(documents=texts, embedding= embedding_model)
|
|
|
71 |
return "Ready..."
|
72 |
|
73 |
|
74 |
+
def process_text_document(document_file):
|
75 |
+
loader = TextLoader(document_file.name)
|
76 |
document = loader.load()
|
77 |
return document
|
78 |
|
79 |
|
80 |
+
def process_csv_document(document_file):
|
81 |
+
loader = CSVLoader(file_path=document_file.name)
|
82 |
document = loader.load()
|
83 |
return document
|
84 |
|
85 |
|
86 |
+
def process_word_document(document_file):
|
87 |
+
loader = UnstructuredWordDocumentLoader(file_path=document_file.name)
|
88 |
document = loader.load()
|
89 |
return document
|
90 |
|