Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Appearance settings
Permalink

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also or learn more about diff comparisons.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also . Learn more about diff comparisons here.
base repository: NVIDIA/DeepLearningExamples
Failed to load repositories. Confirm that selected base ref is valid, then try again.
Loading
base: master
Choose a base ref
Loading
...
head repository: desertfire/DeepLearningExamples
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: master
Choose a head ref
Loading
Checking mergeability… Don’t worry, you can still create the pull request.
  • 10 commits
  • 4 files changed
  • 2 contributors

Commits on Feb 11, 2022

  1. Add a --ltc option to run Lazy Tensor 

    Test Plan: On AWS, `python ./launch.py --model resnet50 --precision AMP --mode benchmark_training --platform DGXA100 /data/datasets01/imagenet_full_size/061417/ --raport-file benchmark.json --epochs 1 --prof 100 --ltc`
    @desertfire
    desertfire committed Feb 11, 2022
    Configuration menu
    Copy the full SHA
    e2ad817 View commit details
    Browse the repository at this point in the history

Commits on Feb 16, 2022

  1. Enable DDP with LazyTensor

    Jiewen Tan committed Feb 16, 2022
    Configuration menu
    Copy the full SHA
    b526e50 View commit details
    Browse the repository at this point in the history

Commits on Feb 17, 2022

  1. - Make DataLoader running in eager instead of lazy 

    - Comment out `loss=reduced_loss.item()`
    @desertfire
    desertfire committed Feb 17, 2022
    Configuration menu
    Copy the full SHA
    2b7e27e View commit details
    Browse the repository at this point in the history

Commits on Feb 18, 2022

  1. Move the CUDA sync and LTC sync at each iteration to the end of itera… 

    …tions

TODO: Each iteration still dump a log of empty information. Need to be
fixed.
    @desertfire
    desertfire committed Feb 18, 2022
    Configuration menu
    Copy the full SHA
    2bd3037 View commit details
    Browse the repository at this point in the history

Commits on Feb 23, 2022

  1. Add --ltc_summary for dumping LTC summary

    @desertfire
    desertfire committed Feb 23, 2022
    Configuration menu
    Copy the full SHA
    cda387e View commit details
    Browse the repository at this point in the history

  2. Move _ltc_init_ts_backend() to after "LTC_TS_CUDA" is set 

    Summary: If we regitster LTC backend with LTC_TS_CUDA as 0, the default
backend device will be CPU and the whole thing will be slow as snale.
    @desertfire
    desertfire committed Feb 23, 2022
    Configuration menu
    Copy the full SHA
    37f9830 View commit details
    Browse the repository at this point in the history

Commits on Feb 24, 2022

  1. Add --sync_every_iter to turn on synchronization at the end of each i… 

    …teration
    @desertfire
    desertfire committed Feb 24, 2022
    Configuration menu
    Copy the full SHA
    2ebcb4c View commit details
    Browse the repository at this point in the history

  2. Add --fuser for specifying which fuser to use

    @desertfire
    desertfire committed Feb 24, 2022
    Configuration menu
    Copy the full SHA
    8be71cb View commit details
    Browse the repository at this point in the history

  3. Pass gradient_as_bucket_view=True to DDP to fix the NaN problem

    Jiewen Tan committed Feb 24, 2022
    Configuration menu
    Copy the full SHA
    a36f4ed View commit details
    Browse the repository at this point in the history

Commits on Feb 25, 2022

  1. Add a --warmup_iters option to specify how many iterations are used … 

    …as warmup
    @desertfire
    desertfire committed Feb 25, 2022
    Configuration menu
    Copy the full SHA
    a4b93bd View commit details
    Browse the repository at this point in the history
Loading
Morty Proxy This is a proxified and sanitized view of the page, visit original site.