Update app.py
Browse files
app.py
CHANGED
|
@@ -24,19 +24,14 @@ print("-----------")
|
|
| 24 |
text_splitter = CharacterTextSplitter(chunk_size=100, chunk_overlap=0)
|
| 25 |
vdocuments = text_splitter.split_documents(documents)
|
| 26 |
|
| 27 |
-
|
| 28 |
# Extract the embedding arrays from the PDF documents
|
| 29 |
embeddings = []
|
| 30 |
-
for doc in
|
| 31 |
-
embeddings.
|
| 32 |
-
|
| 33 |
|
| 34 |
# Create Chroma vector store for API embeddings
|
| 35 |
api_db = Chroma.from_documents(vdocuments, HfApiEmbeddingRetriever, collection_name="api-collection")
|
| 36 |
|
| 37 |
-
#api_db = Chroma.from_texts(embeddings, api_hf_embeddings, collection_name="api-collection")
|
| 38 |
-
|
| 39 |
-
|
| 40 |
# Define the PDF retrieval function
|
| 41 |
def pdf_retrieval(query):
|
| 42 |
# Run the query through the retriever
|
|
@@ -55,4 +50,4 @@ api_tool = gr.Interface(
|
|
| 55 |
)
|
| 56 |
|
| 57 |
# Launch the Gradio interface
|
| 58 |
-
|
|
|
|
| 24 |
text_splitter = CharacterTextSplitter(chunk_size=100, chunk_overlap=0)
|
| 25 |
vdocuments = text_splitter.split_documents(documents)
|
| 26 |
|
|
|
|
| 27 |
# Extract the embedding arrays from the PDF documents
|
| 28 |
embeddings = []
|
| 29 |
+
for doc in vdocuments:
|
| 30 |
+
embeddings.extend(api_hf_embeddings.get_embeddings(doc))
|
|
|
|
| 31 |
|
| 32 |
# Create Chroma vector store for API embeddings
|
| 33 |
api_db = Chroma.from_documents(vdocuments, HfApiEmbeddingRetriever, collection_name="api-collection")
|
| 34 |
|
|
|
|
|
|
|
|
|
|
| 35 |
# Define the PDF retrieval function
|
| 36 |
def pdf_retrieval(query):
|
| 37 |
# Run the query through the retriever
|
|
|
|
| 50 |
)
|
| 51 |
|
| 52 |
# Launch the Gradio interface
|
| 53 |
+
api_tool.launch()
|