Skip to main content

Cross-GPU performance predictions for PyTorch neural network training.

Project description

DeepView.Predict

License Maintainability Test Coverage

A Runtime-Based Computational Performance Predictor for Deep Neural Network Training

DeepView.Predict is a tool that predicts a deep neural network's training iteration execution time on a given GPU. It currently supports PyTorch. To learn more about how DeepView.Predict works, please see our research paper.

Installation

To run DeepView.Predict, you need:

Currently, we have predictors for the following Nvidia GPUs:

GPU Generation Memory Mem. Type SMs
P4000 Pascal 8 GB GDDR5 14
P100 Pascal 16 GB HBM2 56
V100 Volta 16 GB HBM2 80
RTX 2070 Turing 8 GB GDDR6 36
RTX 2080Ti Turing 11 GB GDDR6 68
T4 Turing 16 GB GDDR6 40
RTX 3090 Ampere 24 GB GDDR6X 82
A100 Ampere 40 GB HBM2 108
A40 Ampere 48 GB GDDR6 84
RTX A4000 Ampere 16 GB GDDR6 48
RTX 4000 Turing 8 GB GDDR6 36

Building locally

Installing from pip

Install via pip with the following command

pip install deepview-predict

Installing from source

  1. Install CUPTI

CUPTI is a profiling interface required by DeepView.Predict. Select your version of CUDA here and follow the instructions to add NVIDIA's repository. Then, install CUPTI with:

sudo apt-get install cuda-cupti-xx-x

where xx-x represents the version of CUDA you have installed.

Alternatively, if you do not have root access on your machine, you can use conda to install CUPTI. Select your version of CUDA here and follow the instructions. For example if you have CUDA 11.6.0, you can install CUPTI with:

conda install -c "nvidia/label/cuda-11.6.0" cuda-cupti

After installing CUPTI, add $CONDA_HOME/extras/CUPTI/lib64/ to LD_LIBRARY_PATH to ensure the library is linked.

  1. Install CMake 3.17+.

    • Note that CMake 3.24.0 and 3.24.1 has a bug that breaks DeepView.Predict as it is not able to find the CUPTI directory and you should not use those versions

    • Run the following commands to download and install a precompiled version of CMake 3.24.2

      wget https://github.com/Kitware/CMake/releases/download/v3.24.2/cmake-3.24.2-linux-x86_64.sh
      chmod +x cmake-3.24.2-linux-x86_64.sh
      mkdir /opt/cmake
      sh cmake-3.24.2-linux-x86_64.sh --prefix=/opt/cmake --skip-license
      ln -s /opt/cmake/bin/cmake /usr/local/bin/cmake
      
    • You can verify the version of CMake you installed with the following command

      cmake --version
      
  2. Install Git Large File Storage

  3. Clone the DeepView.Predict package

    git clone https://github.com/CentML/DeepView.Predict
    cd DeepView.Predict
    
  4. Get the pre-trained models used by DeepView.Predict

    git submodule init && git submodule update
    git lfs pull
    
  5. Finally build DeepView.Predict with the following command

    ./analyzer/install-dev.sh
    

Building with Docker

DeepView.Predict has been tested to work on the latest version of NVIDIA NGC PyTorch containers.

  1. To build DeepView.Predict with Docker, first run the NGC container where
docker run --gpus all -it --rm nvcr.io/nvidia/pytorch:XX.XX-py3
  1. Inside the container, clone the repository then build and install DeepView.Predict Python package:
git clone --recursive https://github.com/CentML/DeepView.Predict
./habitat/analyzer/install-dev.sh

Note: DeepView.Predict needs access to your GPU's performance counters, which requires special permissions if you are running with a recent driver (418.43 or later). If you encounter a CUPTI_ERROR_INSUFFICIENT_PRIVILEGES error when running DeepView.Predict, please follow the instructions here and in issue #5.

Usage example

You can verify your DeepView.Predict installation by running the simple usage example:

# example.py
import habitat
import torch
import torchvision.models as models

# Define model and sample inputs
model = models.resnet50().cuda()
image = torch.rand(8, 3, 224, 224).cuda()

# Measure a single inference
tracker = habitat.OperationTracker(device=habitat.Device.RTX2080Ti)
with tracker.track():
    out = model(image)

trace = tracker.get_tracked_trace()
print("Run time on source:", trace.run_time_ms)

# Perform prediction to a single target device
pred = trace.to_device(habitat.Device.V100)
print("Predicted time on V100:", pred.run_time_ms)
python3 example.py

See experiments/run_experiment.py for other examples of DeepView.Predict usage.

Release History

See Releases

License

The code in this repository is licensed under the Apache 2.0 license (see LICENSE and NOTICE), with the exception of the files mentioned below.

This software contains source code provided by NVIDIA Corporation. These files are:

  • The code under cpp/external/cupti_profilerhost_util/ (CUPTI sample code)
  • cpp/src/cuda/cuda_occupancy.h

The code mentioned above is licensed under the NVIDIA Software Development Kit End User License Agreement.

We include the implementations of several deep neural networks under experiments/ for our evaluation. These implementations are copyrighted by their original authors and carry their original licenses. Please see the corresponding README files and license files inside the subdirectories for more information.

Research Paper

DeepView.Predict began as a research project in the EcoSystem Group at the University of Toronto. The accompanying research paper appeared in the proceedings of USENIX ATC'21. If you are interested, you can read a preprint of the paper here.

If you use DeepView.Predict in your research, please consider citing our paper:

@inproceedings{habitat-yu21,
  author = {Yu, Geoffrey X. and Gao, Yubo and Golikov, Pavel and Pekhimenko,
    Gennady},
  title = {{Habitat: A Runtime-Based Computational Performance Predictor for
    Deep Neural Network Training}},
  booktitle = {{Proceedings of the 2021 USENIX Annual Technical Conference
    (USENIX ATC'21)}},
  year = {2021},
}

Contributing

Check out CONTRIBUTING.md for more information on how to help with Habitat.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distributions

deepview_predict-0.1.4-py311-none-any.whl (135.8 MB view hashes)

Uploaded Python 3.11

deepview_predict-0.1.4-py310-none-any.whl (135.7 MB view hashes)

Uploaded Python 3.10

deepview_predict-0.1.4-py39-none-any.whl (135.5 MB view hashes)

Uploaded Python 3.9

deepview_predict-0.1.4-py38-none-any.whl (135.4 MB view hashes)

Uploaded Python 3.8

deepview_predict-0.1.4-py37-none-any.whl (135.3 MB view hashes)

Uploaded Python 3.7

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page