I have a resume data and i want to segment the profile, name, contact, email from the resume. I had perform the spancat annotations from prodigy. and followed the tutorials as in this video
I m confused how to inference the spancat models and what to do next as that is not mentioned in the videos. or can you guide me what to do next ? Yes i have the train.spacy and dev.spacy files.
NOTE : I had a resume data as:
{"text": "Name: helen helen E-Mail: helen.helen@gmail.com Address: Hong Kong, Hong Kong Github: https://github.com/helen LinkedIn: https://linkedin.com/helen Phone No. 192094070156"}
This is the data and how i annotate i will mention as : the whole above text is labelled as : PROFILE
, and othere as the given labels: PROFILE_NAME,PROFILE_EMAIL,PROFILE_ADDRESS,PROFILE_PHONE
Is am i doing something wrong here? Please guide me Appreciated
hi @kushal_pythonist,
By inference, do you mean training?
Since you have the spacy binary files, I presume you ran data-to-spacy
like in the docs:
$ prodigy data-to-spacy ./corpus --spancat covid_articles
To use this data for training with spaCy, you can run:
python -m spacy train ./corpus/config.cfg --paths.train ./corpus/train.spacy --paths.dev ./corpus/dev.spacy
Like the output states, the next step is to run spacy train
. Did you do that?
Can you describe what steps are confusing?
Well i need to check whether the trained spancat model is performing good or bad
yes
yes i have done that but i'm confused what to do next and have no any insights to go further. Can you enlighten me .
I have trained the model using this command. And got stuck and confused
what to do next. Can @kushal_pythonist and @ryanwesslen help me through this and provide guidance. ATM I'm I m null about proceeding further. Thanks