Replace Juno with ProgressLogging #1643
Conversation
|
You could use https://github.com/pfitzseb/TreeViews.jl instead, but I'd recommend just deleting that code, tbh. |
|
The test errors shouldn't appear in practice, but the following should work for iterators that don't have n = length(data)
@withprogress for (i, d) in enumerate(data)
try
gs = gradient(ps) do
loss(batchmemaybe(d)...)
end
update!(opt, ps, gs)
cb()
catch ex
if ex isa StopException
break
elseif ex isa SkipException
continue
else
rethrow(ex)
end
end
@logprogress i / n
end |
|
That assumes the length of the dataset is known already iiuc. Iteration does not. |
|
True, but what would you log then? There are some issues about open ended logging on ProgressLogging.jl, but AFAIK the package only supports finite length logging. |
|
Maybe we don't call |
Updating docs Co-authored-by: Kyle Daruwalla <[email protected]>
Updating docs Co-authored-by: Kyle Daruwalla <[email protected]>
Updating docs Co-authored-by: Kyle Daruwalla <[email protected]>
Updating docs Co-authored-by: Kyle Daruwalla <[email protected]>
Co-authored-by: Kyle Daruwalla <[email protected]>
1675: Adding GRUv3 support. r=darsnack a=mkschleg As per the starting discussion in #1671, we should provide support for variations on the GRU and LSTM cell. In this PR, I added support for the GRU found in v3 of the [original GRU paper](https://arxiv.org/abs/1406.1078). Current support in Flux is for v1 only. [Tensorflow](https://www.tensorflow.org/api_docs/python/tf/keras/layers/GRU) supports several variations, with this as one of the variations. While the feature is added and usable in this PR, this is only a first pass at a design and could use further iterations. Some questions I have: - Should we have new types for each variation of these cells? (another possibility is through parametric options) - Should we have a shared constructor similar to Tensorflow/Pytorch? (it might make sense to rename the current GRU to GRUv1 if we want to do this). ### PR Checklist - [x] Tests are added - [x] Entry in NEWS.md - [x] Documentation, if applicable - [ ] API changes require approval from a committer (different from the author, if applicable) Co-authored-by: Matthew Schlegel <[email protected]> Co-authored-by: Matthew Schlegel <[email protected]>
Temporary measure to unblock CI while we investigate the root cause. I think `@test_broken` is better than increasing the tolerance, as the latter makes it easier to forget about this one entirely.
Only add PR comment with docs build if the docs label is added
Update train.jl to add a more detailed `train!` docstring
1797: Mark destructure gradient test as broken r=DhairyaLGandhi a=ToucheSir Temporary measure to unblock CI while we investigate the root cause. I think ``@test_broken`` is better than increasing the tolerance, as the latter makes it easier to forget about this one entirely. Co-authored-by: Brian Chen <[email protected]>
1805: Add buildkite step to run on Julia LTS r=DhairyaLGandhi a=DhairyaLGandhi Allow bors to unblock some PRs as well. Co-authored-by: Dhairya Gandhi <[email protected]>
|
If you'd like, I can do the cleanup work for this. |
|
This would need a bit more work than a rebase. ProgressLogging isn't exactly compatible with Juno's, but as long as we have some fallbacks in place, we can merge this. |
|
The only API we're switching over is |
|
superseded by #1807 |
|
There's no guarantee for a known length to the |
cc @pfitzseb @vchuravy
@pfitzseb what would be the equivalent to replace the rendering from Juno?