Hello,
Can you help here?
search_knowledge_base function includes only 5 documents while querying, even though the vector table has 20 documents and 100s of records. How can I include all the files/documents from the table?
Here is the debug statements
DEBUG Getting function search_knowledge_base
DEBUG Running: search_knowledge_base(query=Total Amount Due sum by year group by Week Ending)
DEBUG Getting 5 relevant documents for query: Total Amount Due sum by year group by Week Ending
DEBUG Time to get references: 1.5097s
DEBUG ---------- OpenAI Response Start ----------
DEBUG ============== developer ==============
DEBUG ## Instructions
- Use the local knowledge base from PDF files
- Always include sources in your response
- Use markdown to format your answers.
DEBUG ============== user ==============
DEBUG Add all the Total Amount Due and call that as sum of total amount and group by year, please include all the
rows from the table, Please use Week Ending for any date calculation
DEBUG ============== assistant ==============
Here is the Agent code
agent = Agent(
model=OpenAIChat(id=“gpt-4o-mini”),
knowledge=pdf_knowledge_base,
instructions=[“Use the local knowledge base from PDF files”,
“Always include sources in your response”],
show_tool_calls=True,
debug_mode=True,
use_tools=True,
search_knowledge=True,
markdown=True,
)