We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Updated Terms and model training (markdown)
Removed the TODO section, clarified when terms are used in preprocessing
Updated Terms (proper nouns) and model training (markdown)
Capitalized Paratext, removed draft message at the top of the page
Made some changes to the wording and formatting of the "How are term files constructed" section
Created Terms (proper nouns) and model training (markdown)
Update ORCA web address
Removed double quotes around values where the double quotes don't appear in the config file.
Add note about `\` vs `/` for use_test_set_from parameter.
Updated NMT: Usage (markdown)
Updated Using the Python Debugger (markdown)
Updated Configure a model (markdown)
Provide version number for clearml-agent-slurm
Updated Autoscaler Setup Procedure (markdown)
Added a note to attn_implementation about NLLB incompatability
Removed memory growth
Removed eager execution and memory growth options
Created Autoscaler Setup Procedure (markdown)
Separated the current and previous configurations so that it is easier to copy either one of them.
Add section about using the previous default learning rate configuration
Added instructions for producing multiple translations
Created Producing multiple drafts (markdown)