Why we need to pass test dataset while training?

[nashid]

As per the example, we need to pass the training dataset during training.

From the Readme,

mkdir /tmp/nmt_attention_model

python -m nmt.nmt \
    --attention=scaled_luong \
    --src=vi --tgt=en \
    --vocab_prefix=/tmp/nmt_data/vocab  \
    --train_prefix=/tmp/nmt_data/train \
    --dev_prefix=/tmp/nmt_data/tst2012  \
    --test_prefix=/tmp/nmt_data/tst2013 \
    --out_dir=/tmp/nmt_attention_model \
    --num_train_steps=12000 \
    --steps_per_stats=100 \
    --num_layers=2 \
    --num_units=128 \
    --dropout=0.2 \
    --metrics=bleu

I am wondering why we need to pass the test_prefix during the training phase?