BERT model training impelmentation using 1024 A100 GPUs for MLPerf Training v1.1

Related tags

Deep LearningMLPerf_Training_v1.1
Overview

Pre-trained checkpoint and bert config json file

  1. Location of checkpoint and bert config json file

    This MLCommons members Google Drive location contains these files.

    • TensorFlow checkpoint (tf1_ckpt) containing the pre-trained weights.
    • Config file (bert_config.json) which specifies the hyperparameters of the model.

  2. Checkpoint conversion

python convert_tf_checkpoint.py --tf_checkpoint <path/to/checkpointdir_phase1/model.ckpt-28252.index> --bert_config_path <path/to/checkpointdir_phase1/bert_config.json> --output_checkpoint model.ckpt-28252.pt

Download and preprocess datasets

  1. Download dataset and generate the TFRecords for training data and eval data

    BERT Wikipedia dataset preparation

  2. Convert training data and eval data from TFRecords to HDF5

    TF_INPUT_DIR=<path/to/tfrecord_input_dir> HDF5_OUTPUT_DIR=<path/to/hdf5_output_dir> ./run_trans_tfrecord_to_hdf5.sh

  3. 4bins training data

    We split dataset to enable data-load balacning and it can reduce communication overhead.

    Based on the sequence length distribution, split HDF5 training data into 4 part:

    part 1: 0 < sequence length <= 128

    part 2: 128 < sequence length <= 256

    part 3: 256 < sequence length <= 384

    part 4: 384 < sequence length <= 512

    The output_dir contains 4 sub-directories 128, 256, 384 and 512.

cd cleanup_scripts
python run_split_and_chop_hdf5_files.py --input_dir=<path/to/hdf5_datadir> --output_dir=<path/to/4bins_training_datadir>

Prepare the environment

  • Create a virtualenv and install the required packages:
virtualenv venv -p python3.8.7
source venv/bin/activate
pip install -r requirements.txt

# Install mlperf-logging Python package
git clone https://github.com/mlperf/logging.git mlperf-logging
pip install -e mlperf-logging

# Install apex
git clone https://github.com/NVIDIA/apex.git
cd apex
git reset --hard d06404fecab73f152c6cbb89ac2c2e9b7fc24124
git submodule update --init --recursive
git apply ../patch_for_mlperf_trining_v1.1_by_samsung.patch
pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" --global-option="--distributed_adam" --global-option="--distributed_lamb" --global-option="--bnp" --global-option="--xentropy" --global-option="--fast_layer_norm" --global-option="--deprecated_fused_adam"  --global-option="--fmha"  --global-option="--fast_multihead_attn" ./

# Compile mhalib
cd mhalib
python setup.py build
cp build/lib*/mhalib* ../
  • Other software requirements
Softeware Version
python 3.8.7
pytorch 1.9.1
NCCL 2.9.9
CUDA 11.3.0
cudnn 8.2.1.32
cublas 11.4.2
nvidia driver 470.57.02
mofed version 5.4-1.0.3

Run the model

  1. Set hosts address in run_multinode.sh
export hosts=('192.168.16.1' '192.168.16.2')

  1. Launch the training

    Use the following command to run the config_Samsung_Supercomputer21_DGXA100_128x8x16x1.sh in python virtual environment.

PYTHON=<path/to/python> DGXSYSTEM=Samsung_Supercomputer21_DGXA100_128x8x16x1 INPUT_DIR=<path/to/4bins_training_datadir> EVAL_DIR=<path/to/eval_datadir> CHECKPOINTDIR_PHASE1=<path/to/checkpointdir_phase1> NEXP=10 ./run_multinode.sh

Appendix

Our source code is based on MLPerf BERT v0.7, and all the files newly added and modified are as follows.

File Name Status Description
config_Samsung_Supercomputer21_DGXA100_128x8x16x1.sh Newly added The file contains configurations used for 1024 GPUs experiment.
run_split_and_chop_hdf5_files.py Newly added The file is used for generating 4-bin training data.
mhalib/setup.py Modified The file is modified since CUDA upgraded.
optim/init.py Newly added The file is used as the entrance of "optim" module.
optim/acclip.py Newly added The file implements ACClip optimizer for trial.
optim/madgrad.py Newly added The file implements MADGRAD optimizer for trial.
bind_launch.py Newly added The file is added for BERT training on python environment.
bind_pyt.py Modified The file is modified for the following items.
(1) Log compliance;
(2) Add new NUMA binding.
fmha.py Newly added The file is used for adding FMHA operator (refer to MLPerf v1.0).
mlperf_logger.py Modified The file is modified for log compliance.
modeling.py Modified The file is modified for adding FMHA (refer to MLPerf v1.0).
padding.py Modified The file is modified for adding FMHA (refer to MLPerf v1.0).
README.md Modified It is modified to run Samsung optimized implematation.
requirements.txt Modified The file shows required software version.
run_multinode.sh Newly added The file is startup script about how to run BERT training on python environment
run_pretraining.py Modified The file is modified for the following items.
(1) Load splitting training data;
(2) Add exchange padding feature (refer to MLPerf v1.0);
(3) Add NCCL warmup (refer to MLPerf v1.0);
(4) Add SAIT local/group exchange padding;
(5) Add NCCL warmup for group exchange padding;
(6) Add per-device local gradient clipping before all-reduce;
(7) Add pytorch DDP.
schedulers.py Modified The file is modified for optimizing learning rate scheduler
utils.py Modified The file is modified for the following items.
(1) Add get_optimzer() interface;
(2) Add a batch sampler (SplitRandomSampler) for 4-bin splitting training data.
Owner
SAIT (Samsung Advanced Institute of Technology)
SAIT (Samsung Advanced Institute of Technology)
GitHub Repository
KITTI-360 Annotation Tool is a framework that developed based on python(cherrypy + jinja2 + sqlite3) as the server end and javascript + WebGL as the front end.

KITTI-360 Annotation Tool is a framework that developed based on python(cherrypy + jinja2 + sqlite3) as the server end and javascript + WebGL as the front end.

86 Dec 12, 2022
《Improving Unsupervised Image Clustering With Robust Learning》(2020)

Improving Unsupervised Image Clustering With Robust Learning This repo is the PyTorch codes for "Improving Unsupervised Image Clustering With Robust L

Sungwon Park 129 Dec 27, 2022
PyContinual (An Easy and Extendible Framework for Continual Learning)

PyContinual (An Easy and Extendible Framework for Continual Learning) Easy to Use You can sumply change the baseline, backbone and task, and then read

176 Jan 05, 2023
Translation-equivariant Image Quantizer for Bi-directional Image-Text Generation

Translation-equivariant Image Quantizer for Bi-directional Image-Text Generation Woncheol Shin1, Gyubok Lee1, Jiyoung Lee1, Joonseok Lee2,3, Edward Ch

Woncheol Shin 7 Sep 26, 2022
SphereFace: Deep Hypersphere Embedding for Face Recognition

SphereFace: Deep Hypersphere Embedding for Face Recognition By Weiyang Liu, Yandong Wen, Zhiding Yu, Ming Li, Bhiksha Raj and Le Song License SphereFa

Weiyang Liu 1.5k Dec 29, 2022
A collection of resources and papers on Diffusion Models, a darkhorse in the field of Generative Models

This repository contains a collection of resources and papers on Diffusion Models and Score-based Models. If there are any missing valuable resources

5.1k Jan 08, 2023
HNN: Human (Hollywood) Neural Network

HNN: Human (Hollywood) Neural Network Learn the top 1000 actors on IMDB with your very own low cost, highly parallel, CUDAless biological neural netwo

Madhava Jay 0 Dec 21, 2021
Rainbow is all you need! A step-by-step tutorial from DQN to Rainbow

Do you want a RL agent nicely moving on Atari? Rainbow is all you need! This is a step-by-step tutorial from DQN to Rainbow. Every chapter contains bo

Jinwoo Park (Curt) 1.4k Dec 29, 2022
Vision Transformer and MLP-Mixer Architectures

Vision Transformer and MLP-Mixer Architectures Update (2.7.2021): Added the "When Vision Transformers Outperform ResNets..." paper, and SAM (Sharpness

Google Research 6.4k Jan 04, 2023
Code for "Reconstructing 3D Human Pose by Watching Humans in the Mirror", CVPR 2021 oral

Reconstructing 3D Human Pose by Watching Humans in the Mirror Qi Fang*, Qing Shuai*, Junting Dong, Hujun Bao, Xiaowei Zhou CVPR 2021 Oral The videos a

ZJU3DV 178 Dec 13, 2022
A Deep Reinforcement Learning Framework for Stock Market Trading

DQN-Trading This is a framework based on deep reinforcement learning for stock market trading. This project is the implementation code for the two pap

61 Jan 01, 2023
OpenMMLab Computer Vision Foundation

English | 简体中文 Introduction MMCV is a foundational library for computer vision research and supports many research projects as below: MMCV: OpenMMLab

OpenMMLab 4.6k Jan 09, 2023
Repository for the "Gotta Go Fast When Generating Data with Score-Based Models" paper

Gotta Go Fast When Generating Data with Score-Based Models This repo contains the official implementation for the paper Gotta Go Fast When Generating

Alexia Jolicoeur-Martineau 89 Nov 09, 2022
NUANCED is a user-centric conversational recommendation dataset that contains 5.1k annotated dialogues and 26k high-quality user turns.

NUANCED: Natural Utterance Annotation for Nuanced Conversation with Estimated Distributions Overview NUANCED is a user-centric conversational recommen

Facebook Research 18 Dec 28, 2021
tsai is an open-source deep learning package built on top of Pytorch & fastai focused on state-of-the-art techniques for time series classification, regression and forecasting.

Time series Timeseries Deep Learning Pytorch fastai - State-of-the-art Deep Learning with Time Series and Sequences in Pytorch / fastai

timeseriesAI 2.8k Jan 08, 2023
Few-Shot Object Detection via Association and DIscrimination

Few-Shot Object Detection via Association and DIscrimination Code release of our NeurIPS 2021 paper: Few-Shot Object Detection via Association and DIs

Cao Yuhang 49 Dec 18, 2022
LibFewShot: A Comprehensive Library for Few-shot Learning.

LibFewShot Make few-shot learning easy. Supported Methods Meta MAML(ICML'17) ANIL(ICLR'20) R2D2(ICLR'19) Versa(NeurIPS'18) LEO(ICLR'19) MTL(CVPR'19) M

<a href=[email protected]&L"> 603 Jan 05, 2023
PyImpetus is a Markov Blanket based feature subset selection algorithm that considers features both separately and together as a group in order to provide not just the best set of features but also the best combination of features

PyImpetus PyImpetus is a Markov Blanket based feature selection algorithm that selects a subset of features by considering their performance both indi

Atif Hassan 103 Dec 14, 2022
Implementation of the GVP-Transformer, which was used in the paper "Learning inverse folding from millions of predicted structures" for de novo protein design alongside Alphafold2

GVP Transformer (wip) Implementation of the GVP-Transformer, which was used in the paper Learning inverse folding from millions of predicted structure

Phil Wang 19 May 06, 2022
QueryInst: Parallelly Supervised Mask Query for Instance Segmentation

QueryInst is a simple and effective query based instance segmentation method driven by parallel supervision on dynamic mask heads, which outperforms previous arts in terms of both accuracy and speed.

Hust Visual Learning Team 386 Jan 08, 2023