issue facing in prodigy v1.10

Hi, I am shubham jain and currently i am using prodigy v.1.10 . I have use rel. manual syntax for tagging relationship.
!PRODIGY_PORT=8093 prodigy rel.manual newcheck ./org_fa_v1/ ./org_fa_tom.jsonl --label mapping --span-label FinancialAdvisor,ORG --wrap
With this syntax, i was able to mark relationship between two tokens.


like this.

and after annotation i train model with this syntax
!PRODIGY_PORT=8093 prodigy train parser newcheck ./org_fa_v1 --output ./newcheck_model_v2

:heavy_check_mark: Loaded model './org_fa_v1'
Created and merged data for 32 total examples
Using 16 train / 16 eval (split 50%)
Component: parser | Batch size: compounding | Dropout: 0.2 | Iterations: 10
:information_source: Baseline accuracy: 0.000

=========================== :sparkles: Training the model ===========================

:heavy_check_mark: Saved model: /notebooks/shubham/new_start/newcheck_model_v2

I want to ask whether my model is train or not? if it is properly train, then how i will use this custom dep?. can you tell me the syntax?

Hi! I think you might be facing the same problem as described here:

I think there might be another problem that somehow causes it to not extract the dependencies correctly – I'll take a look and we'll include a fix in the upcoming v1.10.2!

In the meantime, you could use Prodigy's db-out to export your annotations and then train with spaCy (or any other library) manually. The data will include the token indices, the labels and the heads, which is all you need to train a dependency parser.