Error : 'name 's' is not defined' while running the PDF

Short Course Q&A Preprocessing Unstructured Data 4 LLM Applications
1

On running the line below for processing PDFs -
resp = s.general.partition(req)

I get the below error-
NameError: name ‘s’ is not defined

Does anyone know how to solve this?

1 Like
2

Make sure that the variable ‘s’ is properly defined and assigned a value before calling the partition method. Also, check for any misspellings or incorrect variable scopes (such as being defined within a different function or block of code) too.

1 Like
3

I am using the code directly as given in the course notebook. I wonder if there is anything that I need to change for it to run successfully.

The code that I am trying to run -
filename = “example_files/CoT.pdf”
with open(filename, “rb”) as f:
files=shared.Files(
content=f.read(),
file_name=filename,
)

req = shared.PartitionParameters(
files=files,
strategy=‘hi_res’,
pdf_infer_table_structure=True,
languages=[“eng”],
)
try:
resp = s.general.partition(req)
print(json.dumps(resp.elements[:3], indent=2))
except SDKError as e:
print(e)

1 Like
4

You must import the required library and define ‘s’ using the UnstructuredClient class from the unstructured_client module:

from unstructured_client import UnstructuredClient
s = UnstructuredClient(api_key_auth="YOUR_API_KEY")

Then, replace “YOUR_API_KEY” with the actual API key obtained from the website https://unstructured.io/

Documentation link: https://unstructured-io.github.io/unstructured/apis/api_sdks.html

3 Likes
5

Thanks Alireza.
This worked for me!

6

You’re welcome! Happy to help :raised_hands:

8

Also worked for me. Thanks!

1 Like