priyakasimbeg

I'm having a hard time loading the jxr into Python. I've tried imageio(with freeimage plugin) as well as imagemagik (wand). Could you please let me know what package can be...

Hi we're planning on discussing this feature request on Th 9/5 during the WG meeting

Added self-reporting instructions in getting started and script to package logs in https://github.com/mlcommons/algorithmic-efficiency/pull/657. I suppose one remaining question is what medium we want to use to download logs?

Perhaps we can use the MLCommons Benchmarking Infra submission tool to allow submitters to upload the logs to as well.

The dataset layouts and final sizes are documented in [datasets/README.md](https://github.com/mlcommons/algorithmic-efficiency/blob/main/datasets/README.md) in the dropdown items saying "The final directory structure should look like this:".

@chandramouli-sastry could you help close this request? I have all the data from the setup scripts downloaded in kasimbeg-8 in /home/kasimbeg/data. The remaining work is to: 1. Check the README...

Changed default number of workers for PyTorch data loaders to 0. Important update: for speech workloads the `pytorch_eval_num_workers` flag to submission_runner.py has to be set to >0, to prevent data...

What is the version of the GPU driver? Can you print the output of nvidia-smi in your environment? We had a similar issue a few months ago https://github.com/mlcommons/algorithmic-efficiency/issues/497

Ok I see, sorry I did not catch that it was consistently OOMing. We can try to clear the cache between trials. Could you try adding `_reset_cuda_mem()` in `submission_runner.py`, line...

Ok thanks for the update. I will make a fix to reset the cuda mem in our repo. We will try to reproduce and investigate the dropout issue on our...