File size: 1,088 Bytes
d1396f0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 |
#!/bin/bash
#SBATCH --job-name=pile-jsonl-to-meg-alpha # job name
#SBATCH --ntasks=1 # number of MP tasks
#SBATCH --nodes=1
#SBATCH --cpus-per-task=40 # number of cores per tasks
#SBATCH --hint=nomultithread # we get physical cores not logical
#SBATCH --time=100:00:00 # maximum execution time (HH:MM:SS)
#SBATCH --qos=qos_cpu-t4
#SBATCH --output=%x-%j.out # output file name
#SBATCH --account=six@cpu
#SBATCH --partition=cpu_p1
set -x -e
source $six_ALL_CCFRWORK/start-prod
export HF_DATASETS_OFFLINE=1
export TRANSFORMERS_OFFLINE=1
input=$six_ALL_CCFRSCRATCH/datasets-custom/pile/train/pile_train.jsonl
output=$six_ALL_CCFRSCRATCH/datasets-custom/pile/train/pile_train_alpha
cd $six_ALL_CCFRWORK/code/Megatron-DeepSpeed
/usr/bin/time -v python tools/preprocess_data.py \
--input $input \
--output-prefix $output \
--dataset-impl mmap \
--tokenizer-type PretrainedFromHF \
--tokenizer-name-or-path bigscience/oscar_13_languages_alpha_weight \
--append-eod \
--workers 25
done
|