Change opuscleaner-mode to 'customs' when generating a config (#1370)

Author: ZJaume

docs/data-and-cleaning/index.md

@@ -41,7 +41,7 @@ See examples in the directory.
 
 ### Default configs
 
-Set `opuscleaner-mode: custom` in the training config to use custom per-dataset and per-language pair configs.
+Set `opuscleaner-mode: custom` (this is the default when generating a config) in the training config to use custom per-dataset and per-language pair configs.
 
 If no custom config was specified for the dataset, 
 the [default config template](https://github.com/mozilla/translations/tree/main/pipeline/clean/opuscleaner/configs/default.filters.json) will be used.
@@ -58,6 +58,9 @@ The config is chosen based on this search order:
 
 The first found config will be applied.
 
+If the desired behaviour is to apply only the default config template and skip all possible custom configs
+for the current language pair and/or datasets, set `opuscleaner-mode: defaults`.
+
 ## Bicleaner
 
 It is recommended to use Bicleaner ML models to filter noisy data.

taskcluster/configs/config.prod.yml

@@ -25,7 +25,9 @@ experiment:
   # Use the Opus Cleaner tool on the data cleaning step.
   # https://github.com/hplt-project/OpusCleaner
   # "custom" to use dataset specific configs, "defaults" to use the same default setting for all datasets
-  opuscleaner-mode: "defaults"
+  # been using "defaults" for some time, but now "custom" will be used
+  # as all the custom filters in the pipeline are made to be used in future retrainings
+  opuscleaner-mode: "custom"
 
   # Archive corpora from alignments tasks to GCS
   archive-corpora: true