Issues
- 0
Parse front and back matter and extra books according to their stylesheets.
#483 opened by davidbaines - 0
Create initial xri_etl script
#473 opened by rminsil - 2
Preprocessing: SettingWithCopyWarning messages when using score_threshold for filtering training pairs
#484 opened by mmartin9684-sil - 0
Backup for S3 bucket
#482 opened by mshannon-sil - 0
- 7
ValueError: Some books failed to translate
#444 opened by jcuenod - 0
- 0
Preprocessing should throw an error / warning if use_test_set_from does not specify a valid experiment folder
#454 opened by mmartin9684-sil - 2
Compare NLLB and MADLAD-400
#398 opened by isaac091 - 4
Extract functions not extracting properly
#466 opened by SirBac0n - 0
(Parent) XRI to extract ETL script
#472 opened by rminsil - 0
- 0
- 0
Onboarding: Select multiple texts after searching
#475 opened by Enkidu93 - 2
Update to use Machine.py for USFM generation
#449 opened by ddaspit - 0
Adjust default learning rate for MADLAD
#418 opened by isaac091 - 3
- 1
Add script to collect verse counts and alignment scores for the projects in an experiment
#425 opened by isaac091 - 3
Unclosed \fe breaks usfm parser
#467 opened by jcuenod - 0
Filename with [ and ] characters being used as regex pattern and failing to match
#464 opened by mshannon-sil - 0
Reduce file size of LoRA models
#397 opened by isaac091 - 1
Onboarding App: Difficulty uploading new projects
#411 opened by Enkidu93 - 0
- 0
- 0
Onboarding: Make sure the Onboarding App is using service credentials for ClearML
#459 opened by Enkidu93 - 2
Translate script doesn't support USFM file names using the "<nn>-<book>" naming format
#456 opened by mmartin9684-sil - 0
- 0
- 0
- 0
Onboarding: Verify that commands/configs used for running jobs are up-to-date with EITL practice
#457 opened by Enkidu93 - 4
Onboarding App: Error when running alignments
#405 opened by Enkidu93 - 0
Onboarding App: Fetching resource list is too slow
#407 opened by Enkidu93 - 0
Onboarding App: Target-text-guessing is incorrect
#413 opened by Enkidu93 - 2
Onboarding App: Auth timeout
#414 opened by Enkidu93 - 0
Onboarding App: Success messages/responsiveness
#412 opened by Enkidu93 - 0
- 0
- 0
Update eflomal to 2.0.0
#431 opened by mshannon-sil - 0
Major KBT's: Option to add them to training data when the source glosses are in a different language than the training source text.
#441 opened by mmartin9684-sil - 0
KBT's: option for a book-based filter for the KBT's added to the training data.
#442 opened by mmartin9684-sil - 0
- 1
- 0
Provide a method for training models where the text is not already known by NLLB.
#422 opened by davidbaines - 0
Research methods of adding tags during training and using them during translation.
#423 opened by davidbaines - 1
Investigate seemingly inaccurate ClearML stats
#419 opened by isaac091 - 0
Improve extraction where an opening marker absorbs the following whitespace.
#420 opened by davidbaines - 0
Use SILNLP docker image for experiments
#416 opened by mshannon-sil - 0
Incorrect caching of results
#406 opened by Enkidu93 - 0
- 1
Automated aligned translation candidates
#402 opened by jcuenod