Closed
Description
It seems that the files in the resources
folder being uploaded are binary files. However, when you prompt (how many countries have population more than usa
) the llm to answer question for it, it is expecting a text file.
Output from LLM:
# AuthorRole.ASSISTANT: I am currently unable to access the content of the uploaded files due to an unexpected issue. Please try re-uploading the files or converting them to a different format, such as Excel or CSV, to see if that resolves the issue.
If I change the file open call from 'rb'
to 'r'
, i get this error:
with open(path, "r") as file:
file = await client.files.create(file=file, purpose="assistants")
Error:
File "C:\r\learn\semantic-kernel\.venv\Lib\site-packages\httpx\_multipart.py", line 163, in __init__
raise TypeError(
"Multipart file uploads must be opened in binary mode, not text mode."
)
TypeError: Multipart file uploads must be opened in binary mode, not text mode.
Am I missing something?
Metadata
Metadata
Assignees
Type
Projects
Status
No status