How to train model

(prodigy_env) D:\AI_projects\ExtractSign>python -m prodigy train image signature_annotation D:\AI_projects\ExtractSign\annotations.jsonl --label "signature"
:information_source: Using CPU

✘ Invalid config override 'signature_annotation': name should start
with --
when I am using this command this kind of error will be there

Hi there!

I'll need a bit more context to help me understand what is happening here. What kind of annotations do you have and what kind of model would you like to train? Could you share what annotation interface you used and what task you had in mind?

The train command expects a reference to a dataset name and it seems that you're passing it a .jsonl file.

what kind of model would you like to train?
Model I want to train is to signature detection from various cards
What kind of annotations do you have?
train a model for image signature annotation. Prodigy is an annotation tool by Explosion AI, which is used to create labeled datasets for machine learning tasks.
image.manual annotation interface I am using

Ah, in that case you'll need to write your own training loop.

The train recipe is able to bootstrap spaCy pipelines on for NLP tasks, but it does not support models for image classification.