From 36e56b7bdb7e0fc4012f7d1d9876895085312976 Mon Sep 17 00:00:00 2001 From: Evgeny Pavlov Date: Wed, 20 Mar 2024 14:45:46 -0700 Subject: [PATCH] Fix pretraining (#485) --- pipeline/train/configs/opustrainer/teacher.yml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/pipeline/train/configs/opustrainer/teacher.yml b/pipeline/train/configs/opustrainer/teacher.yml index d4426c4f..fd168115 100644 --- a/pipeline/train/configs/opustrainer/teacher.yml +++ b/pipeline/train/configs/opustrainer/teacher.yml @@ -6,11 +6,11 @@ stages: - pretrain - finetune -# Back-translated corpus can vary a lot in size, so we can try using original one to count epochs +# Train until the model sees two epochs of back-translated corpus pretrain: - original 0.6 - backtranslated 0.4 - - until original 2 + - until backtranslated 2 # Fine-tuning only on original clean corpus until the early stopping finetune: