You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
ColossalAI/examples/tutorial/sequence_parallel
Frank Lee 8b7495dd54
[example] integrate seq-parallel tutorial with CI (#2463)
2 years ago
..
data [tutorial] added missing dummy dataloader (#1944) 2 years ago
loss_func [tutorial] edited hands-on practices (#1899) 2 years ago
lr_scheduler [tutorial] edited hands-on practices (#1899) 2 years ago
model [tutorial] edited hands-on practices (#1899) 2 years ago
README.md [example] integrate seq-parallel tutorial with CI (#2463) 2 years ago
config.py [example] integrate seq-parallel tutorial with CI (#2463) 2 years ago
requirements.txt [example] integrate seq-parallel tutorial with CI (#2463) 2 years ago
test_ci.sh [example] integrate seq-parallel tutorial with CI (#2463) 2 years ago
train.py [example] integrate seq-parallel tutorial with CI (#2463) 2 years ago

README.md

Sequence Parallelism

Table of contents

📚 Overview

In this tutorial, we implemented BERT with sequence parallelism. Sequence parallelism splits the input tensor and intermediate activation along the sequence dimension. This method can achieve better memory efficiency and allows us to train with larger batch size and longer sequence length.

Paper: Sequence Parallelism: Long Sequence Training from System Perspective

🚀 Quick Start

  1. Install PyTorch

  2. Install the dependencies.

pip install -r requirements.txt
  1. Run with the following command
export PYTHONPATH=$PWD

# run with synthetic dataset
colossalai run --nproc_per_node 4 train.py

The default config is sequence parallel size = 2, pipeline size = 1, lets change pipeline size to be 2 and try it again.

🏎 How to Train with Sequence Parallelism

We provided train.py for you to execute training. Before invoking the script, there are several steps to perform.

Step 1. Configure your parameters

In the config.py provided, a set of parameters are defined including training scheme, model, etc. You can also modify the ColossalAI setting. For example, if you wish to parallelize over the sequence dimension on 8 GPUs. You can change size=4 to size=8. If you wish to use pipeline parallelism, you can set pipeline=<num_of_pipeline_stages>.

Step 2. Invoke parallel training

Lastly, you can start training with sequence parallelism. How you invoke train.py depends on your machine setting.

  • If you are using a single machine with multiple GPUs, PyTorch launch utility can easily let you start your script. A sample command is like below:

      colossalai run --nproc_per_node <num_gpus_on_this_machine> --master_addr localhost --master_port 29500 train.py
    
  • If you are using multiple machines with multiple GPUs, we suggest that you refer to colossalai launch_from_slurm or colossalai.launch_from_openmpi as it is easier to use SLURM and OpenMPI to start multiple processes over multiple nodes. If you have your own launcher, you can fall back to the default colossalai.launch function.