single_process_pretrain.sh

#!/bin/bash

# Copyright (c) 2019 NVIDIA CORPORATION. All rights reserved.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

lrank=$OMPI_COMM_WORLD_LOCAL_RANK
comm_rank=$OMPI_COMM_WORLD_RANK
comm_size=$OMPI_COMM_WORLD_SIZE
WORKSPACE="/workspace"

#parameters
train_batch_size=24
learning_rate=${2:-"6e-3"}
precision=${3:-"fp16"}
num_gpus=${4:-4}
warmup_proportion=${5:-"0.2843"}
train_steps=${6:-7038}
save_checkpoint_steps=${7:-200}
resume_training=${8:-"false"}
create_logfile=${9:-"true"}
accumulate_gradients=${10:-"true"}
gradient_accumulation_steps=${11:-4}
seed=${12:-12439}
job_name=${13:-"bert_lamb_pretraining"}
allreduce_post_accumulation=${14:-"true"}
allreduce_post_accumulation_fp16=${15:-"true"}
train_batch_size_phase2=${16:-4096}
learning_rate_phase2=${17:-"4e-3"}
warmup_proportion_phase2=${18:-"0.128"}
train_steps_phase2=${19:-1563}
gradient_accumulation_steps_phase2=${20:-512}
DATASET=${WORKSPACE}/lower_case_1_seq_len_128_max_pred_20_masked_lm_prob_0.15_random_seed_12345_dupe_factor_5_shard_1472_test_split_10/wikicorpus_en/training # change this for other datasets
DATA_DIR_PHASE1=${21:-$BERT_PREP_WORKING_DIR/${DATASET}/}
BERT_CONFIG=${WORKSPACE}/uncased_L-12_H-768_A-12/bert_config.json
CODEDIR=${22:-"`pwd`"}
init_checkpoint=${23:-"None"}
RESULTS_DIR=$CODEDIR/results
CHECKPOINTS_DIR=$RESULTS_DIR/checkpoints

mkdir -p $CHECKPOINTS_DIR


if [ ! -d "$DATA_DIR_PHASE1" ] ; then
   echo "Warning! $DATA_DIR_PHASE1 directory missing. Training cannot start"
fi
if [ ! -d "$RESULTS_DIR" ] ; then
   echo "Error! $RESULTS_DIR directory missing."
   exit -1
fi
if [ ! -d "$CHECKPOINTS_DIR" ] ; then
   echo "Warning! $CHECKPOINTS_DIR directory missing."
   echo "Checkpoints will be written to $RESULTS_DIR instead."
   CHECKPOINTS_DIR=$RESULTS_DIR
fi
if [ ! -f "$BERT_CONFIG" ] ; then
   echo "Error! BERT large configuration file not found at $BERT_CONFIG"
   exit -1
fi

PREC=""
if [ "$precision" = "fp16" ] ; then
   PREC="--fp16"
elif [ "$precision" = "fp32" ] ; then
   PREC=""
elif [ "$precision" = "tf32" ] ; then
   PREC=""
else
   echo "Unknown <precision> argument"
   exit -2
fi

ACCUMULATE_GRADIENTS=""
if [ "$accumulate_gradients" == "true" ] ; then
   ACCUMULATE_GRADIENTS="--gradient_accumulation_steps=$gradient_accumulation_steps"
fi

CHECKPOINT=""
if [ "$resume_training" == "true" ] ; then
   CHECKPOINT="--resume_from_checkpoint"
fi

ALL_REDUCE_POST_ACCUMULATION=""
if [ "$allreduce_post_accumulation" == "true" ] ; then
   ALL_REDUCE_POST_ACCUMULATION="--allreduce_post_accumulation"
fi

ALL_REDUCE_POST_ACCUMULATION_FP16=""
if [ "$allreduce_post_accumulation_fp16" == "true" ] ; then
   ALL_REDUCE_POST_ACCUMULATION_FP16="--allreduce_post_accumulation_fp16"
fi

INIT_CHECKPOINT=""
if [ "$init_checkpoint" != "None" ] ; then
   INIT_CHECKPOINT="--init_checkpoint=$init_checkpoint"
fi

echo $DATA_DIR_PHASE1
INPUT_DIR=$DATA_DIR_PHASE1
CMD=" $CODEDIR/run_pretraining_v1.py"
CMD+=" --input_dir=$DATA_DIR_PHASE1"
CMD+=" --output_dir=$CHECKPOINTS_DIR"
CMD+=" --config_file=$BERT_CONFIG"
CMD+=" --bert_model=bert-large-uncased"
CMD+=" --train_batch_size=$train_batch_size"
CMD+=" --max_seq_length=128"
CMD+=" --max_predictions_per_seq=20"
CMD+=" --max_steps=$train_steps"
CMD+=" --warmup_proportion=$warmup_proportion"
CMD+=" --num_steps_per_checkpoint=$save_checkpoint_steps"
CMD+=" --learning_rate=$learning_rate"
CMD+=" --seed=$seed"
CMD+=" $PREC"
CMD+=" $ACCUMULATE_GRADIENTS"
CMD+=" $CHECKPOINT"
CMD+=" $ALL_REDUCE_POST_ACCUMULATION"
CMD+=" $ALL_REDUCE_POST_ACCUMULATION_FP16"
CMD+=" $INIT_CHECKPOINT"
CMD+=" --do_train"
CMD+=" --json-summary ${RESULTS_DIR}/dllogger.json "
CMD+=" --local_rank ${comm_rank} "
CMD+=" --dist_url tcp://${1}:45679 "
CMD+=" --world_size ${comm_size} "

#CMD="python3 -m torch.distributed.launch --nproc_per_node=$num_gpus $CMD"
APP="python3 $CMD"


if [ "$create_logfile" = "true" ] ; then
  export GBS=$(expr $train_batch_size \* $num_gpus)
  printf -v TAG "pyt_bert_pretraining_phase1_%s_gbs%d" "$precision" $GBS
  DATESTAMP=`date +'%y%m%d%H%M%S'`
  LOGFILE=$RESULTS_DIR/$job_name.$TAG.$DATESTAMP.log
  printf "Logs written to %s\n" "$LOGFILE"
fi

echo ${CMD}

case ${lrank} in
[0])
  export HIP_VISIBLE_DEVICES=0,1,2,3
  export UCX_NET_DEVICES=mlx5_0:1
  export UCX_IB_PCI_BW=mlx5_0:50Gbs
  GLOO_SOCKET_IFNAME=ib0 numactl --cpunodebind=0 --membind=0 ${APP}
  ;;
[1])
  export HIP_VISIBLE_DEVICES=0,1,2,3
  export UCX_NET_DEVICES=mlx5_1:1
  export UCX_IB_PCI_BW=mlx5_1:50Gbs
  GLOO_SOCKET_IFNAME=ib1 numactl --cpunodebind=1 --membind=1 ${APP}
  ;;
[2])
  export HIP_VISIBLE_DEVICES=0,1,2,3
  export UCX_NET_DEVICES=mlx5_2:1
  export UCX_IB_PCI_BW=mlx5_2:50Gbs
  GLOO_SOCKET_IFNAME=ib2 numactl --cpunodebind=2 --membind=2 ${APP}
  ;;
[3])
  export HIP_VISIBLE_DEVICES=0,1,2,3
  export UCX_NET_DEVICES=mlx5_3:1
  export UCX_IB_PCI_BW=mlx5_3:50Gbs
  GLOO_SOCKET_IFNAME=ib3 numactl --cpunodebind=3 --membind=3 ${APP}
  ;;
esac