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.

Update NeptuneLogger and handlers implementations to new API

See original GitHub issue

🚀 Feature

According to the docs : https://docs.neptune.ai/about/legacy/#initialization and https://docs.neptune.ai/about/legacy/#scores-and-metrics current NeptuneLogger implementation is legacy and may fail with new neptune versions.

cc @Blaizzy

Issue Analytics

State:
Created a year ago
Comments:7

Top GitHub Comments

1reaction

Blaizzycommented, Nov 2, 2022

Great news on the dependency conflict with gym

The team had already noticed and fixed it here and will be a part of a next release.

0reactions

Blaizzycommented, Nov 2, 2022

Is it from Neptune side or pytorch-ignite code ?

Neptune side, and we’ll send a PR to your repo once it’s done. 😃

Anyway, can you update our code to the newer API?

We have completed the specification of the update, and now it’s marked for development, but there is no ETA for it.

Top Results From Across the Web

Issues · pytorch/ignite - GitHub

WandBLogger and TensorboardLogger have different APIs for logging audio ... Update NeptuneLogger and handlers implementations to new API enhancement help ...

NeptuneLogger — PyTorch Lightning 1.8.5.post0 documentation

Automatically log checkpointed model. Called after model checkpoint callback saves a new checkpoint. Parameters. checkpoint_callback ( Checkpoint ) – the model ...

PyTorch Lightning vs Ignite: What Are the Differences?

The Lightning API offers the same functionality as raw Pytorch, ... call the NeptuneLogger from ignite.contrib.handlers.neptune_logger.

PyTorch Lightning 1.5 Released - Exxact Corporation

See what's new in the recent update of PyTorch Lightning, version 1.5. ... With the new Lightning Loop API in v1.5, you can...

neptune_logger — PyTorch-Ignite v0.4.10 Documentation

Neptune logger and its helper handlers. Classes. GradsScalarHandler. Helper handler to log model's gradients as scalars.