You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Nov 3, 2023. It is now read-only.
Bug description
When I use following command you provided in projects/hallucination/README.md,
I cannot train retrieval-based models.
When I use the option --m instead of -m, it works.
Reproduction steps
For example, Run the following:
parlai train_model -m rag -t wizard_of_wikipedia
--rag-model-type token --rag-retriever-type dpr --dpr-model-file zoo:hallucination/multiset_dpr/hf_bert_base.cp
--generation-model bart -o arch/bart_large
--retriever-debug-index compressed
--batchsize 2 --fp16 True --gradient-clip 0.1 --label-truncate 128
--log-every-n-secs 30 --lr-scheduler reduceonplateau --lr-scheduler-patience 1
--model-parallel True --optimizer adam --text-truncate 512 --truncate 512
-lr 1e-05 -vmm min -veps 0.25 -vme 1000 -vmt ppl -vp 5
Expected behavior
The options should be overridden by argument provided in command line
Bug description
When I use following command you provided in projects/hallucination/README.md,
I cannot train retrieval-based models.
When I use the option --m instead of -m, it works.
Reproduction steps
For example, Run the following:
parlai train_model -m rag -t wizard_of_wikipedia
--rag-model-type token --rag-retriever-type dpr --dpr-model-file zoo:hallucination/multiset_dpr/hf_bert_base.cp
--generation-model bart -o arch/bart_large
--retriever-debug-index compressed
--batchsize 2 --fp16 True --gradient-clip 0.1 --label-truncate 128
--log-every-n-secs 30 --lr-scheduler reduceonplateau --lr-scheduler-patience 1
--model-parallel True --optimizer adam --text-truncate 512 --truncate 512
-lr 1e-05 -vmm min -veps 0.25 -vme 1000 -vmt ppl -vp 5
Expected behavior
The options should be overridden by argument provided in command line
Logs
Additional context
Add any other context about the problem here. (like proxy settings, network setup, overall goals, etc.)
The text was updated successfully, but these errors were encountered: