Hi Team,
I want to run the full pipeline in AI Center. Can somebody please tell me what needs to be provided in the Evaluation Dataset?
I created one empty folder in AI Center to run the pipe and it failed stating - “Train successful and Eval failed”
Welcome to the community
As the name suggest its for evaluation…
Upload similar kind of documents you uploaded for training dataset…
Try to upload differenr documents than used in training …that way accuracy can be found out…you can do with the same set of documents as well but as it is trained on same documents it might give 100% results
Hope this helps
Cheers
Does this require any specific folder path?
It failed to evaluate again.
Data set path: Test> File name
Hi @ascent_way
Welcome to the Community!
How did you upload the document?
It has to be done using the Document Manager and in there you have to select it as an evaluation se,
and then you could export it to AI Center and then use the folder while creating pipeline.
Thanks
You have to enable the evaluation dataset option in the created dataset
And the folder structures valries depending on mode…but i guess you are doing a po scrapping so should be fine
If not then please let us know…basically I dont think you used any in training too
Cheers
This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.