Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Allow specification for GPU device index #96

Merged
merged 43 commits into from
Mar 28, 2024
Merged

Allow specification for GPU device index #96

merged 43 commits into from
Mar 28, 2024

Commits on Mar 19, 2024

  1. Have get_device use torch::Device

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    83ff00a View commit details
    Browse the repository at this point in the history

  2. Add device_number arg for get_device

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    a392900 View commit details
    Browse the repository at this point in the history

  3. Throw error if device_number used in CPU-only case

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    2552c91 View commit details
    Browse the repository at this point in the history

  4. Disallow negative device number

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    9b0b7dd View commit details
    Browse the repository at this point in the history

  5. Actually use the device number

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    e44e3e6 View commit details
    Browse the repository at this point in the history

  6. Use device number for torch_zeros

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    cf39472 View commit details
    Browse the repository at this point in the history

  7. Use device number for torch_ones

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    01b8063 View commit details
    Browse the repository at this point in the history

  8. Use device number for torch_empty

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    530fa19 View commit details
    Browse the repository at this point in the history

  9. Use device number for torch_from_blob

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    af7a8af View commit details
    Browse the repository at this point in the history

  10. Device and device number args for torch_module_load

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    e2fe070 View commit details
    Browse the repository at this point in the history

  11. Pass device and device number to torch_jit_load by value

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    fd729a3 View commit details
    Browse the repository at this point in the history

  12. Make device number argument to torch_module_load optional

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    3b3e62c View commit details
    Browse the repository at this point in the history

  13. Make device number argument to torch_tensor_from_array optional

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    5fe34b0 View commit details
    Browse the repository at this point in the history

  14. Make device number argument to other subroutines optional

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    3fe5258 View commit details
    Browse the repository at this point in the history

  15. Make device argument to torch_module_load optional

    @jwallwork23
    jwallwork23 committed Mar 19, 2024
    Configuration menu
    Copy the full SHA
    9ed2452 View commit details
    Browse the repository at this point in the history

Commits on Mar 20, 2024

  1. Add function for determining device_index

    @jwallwork23
    jwallwork23 committed Mar 20, 2024
    Configuration menu
    Copy the full SHA
    fbc6a12 View commit details
    Browse the repository at this point in the history

  2. Rename device number as index

    @jwallwork23
    jwallwork23 committed Mar 20, 2024
    Configuration menu
    Copy the full SHA
    58d28ed View commit details
    Browse the repository at this point in the history

  3. Rename device as device type

    @jwallwork23
    jwallwork23 committed Mar 20, 2024
    Configuration menu
    Copy the full SHA
    682d887 View commit details
    Browse the repository at this point in the history

  4. Device index defaults to -1 on CPU and 0 on GPU

    @jwallwork23
    jwallwork23 committed Mar 20, 2024
    Configuration menu
    Copy the full SHA
    2d9698c View commit details
    Browse the repository at this point in the history

  5. Make device type and index optional on C++ side

    @jwallwork23
    jwallwork23 committed Mar 20, 2024
    Configuration menu
    Copy the full SHA
    ca40777 View commit details
    Browse the repository at this point in the history

  6. Fix typo in torch_model_load

    @jwallwork23
    jwallwork23 committed Mar 20, 2024
    Configuration menu
    Copy the full SHA
    e37f743 View commit details
    Browse the repository at this point in the history

Commits on Mar 22, 2024

  1. Fix typos in example 1

    @jwallwork23
    jwallwork23 committed Mar 22, 2024
    Configuration menu
    Copy the full SHA
    8b63dfe View commit details
    Browse the repository at this point in the history

  2. Initial draft of example 3_MultiGPU

    @jwallwork23
    jwallwork23 committed Mar 22, 2024
    Configuration menu
    Copy the full SHA
    8982129 View commit details
    Browse the repository at this point in the history

Commits on Mar 25, 2024

  1. Differentiate between errors and warnings in C++ code

    @jwallwork23
    jwallwork23 committed Mar 25, 2024
    Configuration menu
    Copy the full SHA
    1eec646 View commit details
    Browse the repository at this point in the history

  2. Formatting

    @jwallwork23
    jwallwork23 committed Mar 25, 2024
    Configuration menu
    Copy the full SHA
    2739c16 View commit details
    Browse the repository at this point in the history

  3. Add mpi4py to requirements for example 3

    @jwallwork23
    jwallwork23 committed Mar 25, 2024
    Configuration menu
    Copy the full SHA
    fc18b52 View commit details
    Browse the repository at this point in the history

  4. Use mpi4py to differ inputs in simplenet_infer_python

    @jwallwork23
    jwallwork23 committed Mar 25, 2024
    Configuration menu
    Copy the full SHA
    2b0086a View commit details
    Browse the repository at this point in the history

  5. Raise ValueError for Python inference with invalid device

    @jwallwork23
    jwallwork23 committed Mar 25, 2024
    Configuration menu
    Copy the full SHA
    fced4c1 View commit details
    Browse the repository at this point in the history

  6. Print rank in Python case; updates to README

    @jwallwork23
    jwallwork23 committed Mar 25, 2024
    Configuration menu
    Copy the full SHA
    188b305 View commit details
    Browse the repository at this point in the history

  7. Setup MPI for simplenet_infer_fortran, too

    @jwallwork23
    jwallwork23 committed Mar 25, 2024
    Configuration menu
    Copy the full SHA
    dcfb153 View commit details
    Browse the repository at this point in the history

  8. Write formatting for example 3

    @jwallwork23
    jwallwork23 committed Mar 25, 2024
    Configuration menu
    Copy the full SHA
    392afb9 View commit details
    Browse the repository at this point in the history

  9. Add note on building with Make

    @jwallwork23
    jwallwork23 committed Mar 25, 2024
    Configuration menu
    Copy the full SHA
    9fd3040 View commit details
    Browse the repository at this point in the history

Commits on Mar 27, 2024

  1. Print before and after; mpi_finalise; output on CPU; comments

    @jwallwork23
    jwallwork23 committed Mar 27, 2024
    Configuration menu
    Copy the full SHA
    24d5b6a View commit details
    Browse the repository at this point in the history

  2. Merge branch 'main' into 85_gpu_device_number

    @jwallwork23
    jwallwork23 committed Mar 27, 2024
    Configuration menu
    Copy the full SHA
    a44e262 View commit details
    Browse the repository at this point in the history

  3. Docs: device->device_type for consistency

    @jwallwork23
    jwallwork23 committed Mar 27, 2024
    Configuration menu
    Copy the full SHA
    5ebe845 View commit details
    Browse the repository at this point in the history

  4. Add docs on MultiGPU

    @jwallwork23
    jwallwork23 committed Mar 27, 2024
    Configuration menu
    Copy the full SHA
    18fca7b View commit details
    Browse the repository at this point in the history

Commits on Mar 28, 2024

  1. Update warning text for defaulting to 0 

    Co-authored-by: jatkinson1000 <[email protected]>
    @jwallwork23 @jatkinson1000
    jwallwork23 and jatkinson1000 authored Mar 28, 2024
    Configuration menu
    Copy the full SHA
    475a859 View commit details
    Browse the repository at this point in the history

  2. Mention MPI in requirements

    @jwallwork23
    jwallwork23 committed Mar 28, 2024
    Configuration menu
    Copy the full SHA
    3f26457 View commit details
    Browse the repository at this point in the history

  3. Update outputs for example 3

    @jwallwork23
    jwallwork23 committed Mar 28, 2024
    Configuration menu
    Copy the full SHA
    3dba29a View commit details
    Browse the repository at this point in the history

  4. Use NP rather than 4 GPUs

    @jwallwork23
    jwallwork23 committed Mar 28, 2024
    Configuration menu
    Copy the full SHA
    0e3272e View commit details
    Browse the repository at this point in the history

  5. Implement SimpleNet in example 3 but with a twist

    @jwallwork23
    jwallwork23 committed Mar 28, 2024
    Configuration menu
    Copy the full SHA
    99d3b5b View commit details
    Browse the repository at this point in the history

  6. Add code snippets for multi-GPU doc section

    @jwallwork23
    jwallwork23 committed Mar 28, 2024
    Configuration menu
    Copy the full SHA
    99002d5 View commit details
    Browse the repository at this point in the history

  7. Add note about multiple GPU support to README.md.

    @jatkinson1000
    jatkinson1000 committed Mar 28, 2024
    Configuration menu
    Copy the full SHA
    e2b68bd View commit details
    Browse the repository at this point in the history