question-mark
Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Deprecate TorchRuntimeClient, corresponding classes, and the tutorial

See original GitHub issue

What is the expected enhancement?

The TorchRuntime program does not use primitives, so it will be removed soon. Before that we have to deprecate everything that is related to TorchRuntimeClient. This includes:

  • The runtime package with two scripts, hookbase.py and torch_runtime_client.py
  • The tutorial 06_torch_runtime.ipynb
  • Other components if any

In case of the tutorial, I think, it is enough to add a text in bold in the beginning saying that the tutorial is deprecated, the corresponding runtime programs are deprecated and will be removed.

Issue Analytics

  • State:closed
  • Created a year ago
  • Comments:12 (9 by maintainers)

github_iconTop GitHub Comments

1reaction
1ucian0commented, Sep 29, 2022

Its a min of 3 months which has to include two releases per current Qiskit deprecation policy for removal (item 3 there)

In principle, the depreciation policy applies to the module API (as the code included in the package), not to the service API. I would make sure qiskit-machine-learning code fails gracefully once the service is gone

0reactions
manoelmarquescommented, Sep 29, 2022

So for a minimum deprecation time, I believe the msg would be something like:

....is deprecated as of version 0.5.0 and will be removed next release..."

That would be the shortest period unless a bug fix release is released just to remove it. Then the period could be 1 week, 3 days, one weekend 😃 Another option is to just remove the code altogether and write in the release notes it has been removed. Since it is a service and not an api like mentioned above, the whole service would be gone.

Read more comments on GitHub >

github_iconTop Results From Across the Web

torchrun (Elastic Launch) — PyTorch 1.13 documentation
torchrun supports the same arguments as torch.distributed.launch except for --use_env which is now deprecated. To migrate from torch.distributed.launch to ...
Read more >
Multi-GPU DDP Training with Torchrun (code walkthrough)
In the fourth video of this series, Suraj Subramanian walks through all the code required to implement fault-tolerance in distributed ...
Read more >
Multi node training with PyTorch DDP, torch.distributed.launch ...
Based on the blog post:"Multi-node PyTorch Distributed Training ... training with PyTorch DDP, torch.distributed.launch, torchrun and mpirun.
Read more >
Torchrun command not found for distributed training, does it ...
I was following the torchrun tutorial but at no point were we told how to install torchrun. Should it just be automatically there...
Read more >
Multi node PyTorch Distributed Training Guide For People In A ...
The goal of this tutorial is to give a summary of how to write and launch PyTorch distributed data parallel jobs across multiple...
Read more >

github_iconTop Related Medium Post

No results found

github_iconTop Related StackOverflow Question

No results found

github_iconTroubleshoot Live Code

Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free

github_iconTop Related Reddit Thread

No results found

github_iconTop Related Hackernoon Post

No results found

github_iconTop Related Tweet

No results found

github_iconTop Related Dev.to Post

No results found

github_iconTop Related Hashnode Post

No results found