I am converting a .riva
file to an .rmir
file with the command shown below.
QUESTION: given that program option --gpus 0
is used, do we still need to execute this docker-command on a GPU / NVIDIA server?
docker run --rm --gpus 0 -v $MODEL_LOC:/data $RIVA_SM_CONTAINER --
riva-build speech_recognition
/data/rmir/asr_offline_conformer_ctc.rmir:$KEY
/data/$MODEL_NAME:$KEY
--offline
--name=asr_offline_conformer_ctc_pipeline
--decoder_type=greedy
--ms_per_timestep=40
--chunk_size=4.8
--left_padding_size=1.6
--right_padding_size=1.6
--max_batch_size=16
--nn.fp16_needs_obey_precision_pass
--featurizer.use_utterance_norm_params=False
--featurizer.precalc_norm_time_steps=0
--featurizer.precalc_norm_params=False
--featurizer.max_batch_size=512
--featurizer.max_execution_batch_size=512
--language_code=en-US