mozilla translations issues

mozilla / translations

The code, training pipeline, and models that power Firefox Translations

https://mozilla.github.io/translations/

Mozilla Public License 2.0

154 stars 33 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Publish experiments to W&B from the CI

#817 vrigal closed 2 months ago
2
en-tr translation quality feedback

#816 selimsum opened 2 months ago
9
Experiment with one stage training

#815 eu9ene closed 2 months ago
0
Experiment with data cleaning

#814 eu9ene closed 2 months ago
0
Release cherry picks 2

#813 eu9ene closed 2 months ago
0
Add a train task

#812 gregtatum closed 2 months ago
0
en-hr model does not always correctly distinguish between Indian and Native American

#811 nordzilla opened 2 months ago
0
Taskcluster train actions fail

#810 eu9ene closed 1 month ago
7
Support uploading GCP Taskcluster artifacts to W&B

#809 eu9ene closed 1 month ago
1
Taskcluster evaluation artifacts on GCP are missing an importer

#808 eu9ene opened 2 months ago
2
Bump disk for cefilter

#807 eu9ene closed 2 months ago
0
[taskcluster:error] Error uploading artifact: S3 returned status code 400 which could be an intermittent issue

#806 eu9ene closed 1 month ago
16
Suppress the ruff import sorting behavior

#805 gregtatum closed 2 months ago
0
Explore uploading models to Hugging Face

#804 eu9ene opened 2 months ago
5
[meta] Train low resource languages

#803 gregtatum opened 2 months ago
0
Create a Python package to use translation models

#802 eu9ene opened 2 months ago
3
WIP Patch stack with nllb and hplt importer work

#801 gregtatum closed 1 month ago
0
Make sure square brackets instead of parentheses don't lead to wrong translations

#800 marco-c opened 2 months ago
0
Suffix W&B runs with task group ID for offline Taskcluster publication from GCP

#799 vrigal closed 1 month ago
13
Add an action to rebuild pipeline toolchains and docker images

#798 gabrielBusta closed 1 month ago
1
Change base image to trigger toolchain rebuilds

#797 gabrielBusta closed 2 months ago
0
Change base image to trigger toolchain rebuilds

#796 gabrielBusta closed 2 months ago
1
production and staging repositories share caches

#795 bhearsum opened 2 months ago
0
Change base image to trigger toolchain rebuilds

#794 gabrielBusta closed 2 months ago
1
Perform a comprehensive testing before the final reuploading

#793 eu9ene opened 2 months ago
4
Add publishing to CI

#792 eu9ene closed 2 months ago
4
Change base image to trigger toolchain rebuilds

#791 gabrielBusta closed 2 months ago
1
Filter monolingual synthesized distillation data with a fluency score

#790 gregtatum opened 2 months ago
0
Filter monolingual data based on fluency scores

#789 gregtatum opened 2 months ago
1
Clean up after bicleaner downloader

#788 gregtatum closed 2 months ago
0
Rewrite merge mono and add support for an OPUS monolingual importer

#787 gregtatum closed 2 months ago
0
Bump disk for student

#786 eu9ene closed 2 months ago
0
Improve GPU utilization for "translate" tasks

#785 eu9ene opened 3 months ago
5
Fix restarting downloads

#784 gregtatum closed 3 months ago
0
Improve GPU utilization in student training

#783 eu9ene opened 3 months ago
0
add 2tb gpu workers

#782 bhearsum closed 3 months ago
0
fix: don't run evaluate tasks on pretrained models

#781 bhearsum closed 1 month ago
0
Add a mono nllb build script

#780 gregtatum closed 3 months ago
0
Expand out marian command arguments

#779 gregtatum closed 2 months ago
0
Investigate removing teacher ensemble training

#778 gregtatum opened 3 months ago
0
restrict github-push taskcluster events to `main`

#777 bhearsum closed 1 month ago
8
feat: add scaffolding and basic tests for taskgraph generation

#776 bhearsum closed 2 months ago
0
temp: use cpu worker pool that uses gpu image to see if it works

#775 bhearsum closed 3 months ago
1
train-student OSError: [Errno 28] No space left on device

#774 eu9ene opened 3 months ago
2
Figure out the behavior of OpusTrainer augmentation on student distillation gap

#773 gregtatum closed 3 weeks ago
3
Investigate improving en-lt student distillation by adding more data

#772 gregtatum closed 3 days ago
1
Reduce monolingual data for da-en to investigate distillation performance

#771 gregtatum closed 1 day ago
7
Simplify translate mono kind files

#770 gregtatum closed 3 months ago
0
Normalize punctuation marks in parallel data

#769 gregtatum opened 3 months ago
0
Investigate using LLMs for data augmentation

#768 marco-c opened 3 months ago
1

Previous Next