Training an engine

The training process is the step when the machine learns from the corpora. This can happen when you create a new engine, but you can also retrain your existing engines after certain milestones, for example when a project has finished. The process is similar to the continuous improvement of translation memories (TMs): with every segment added to your engine, it will produce better results.

Training a new engine

1. Upload or import the corpora you want to build your engine from. These are typically TMX files, but you can also use TBX, CSV, TSV, or even translated CAT tool files with confirmed segments.

2. Create a new engine and choose which corpora you want to include.

3. Click the Train button.

When training finishes, the engine is ready for translation, but it needs to be

• published for file-based translation

• deployed for text-based translation

Retraining an existing engine

1. Optional: Update the engine.

2. Click the Train button on the Next version.
   
   

   Open

   

   
   For domain-adapted custom engines where only master corpora have been updated or added, you can choose between two modes of training.

While an engine is being retrained, it can still be used for translation.

Training queue

Engines sent to training may or may not spend some time in the training queue if you are using Globalese in the cloud, depending on the number of engines already queued by other users.