Update changelog for the v0.15 release #1204
Conversation
Codecov Report
@@ Coverage Diff @@
## master #1204 +/- ##
=======================================
Coverage 98.13% 98.13%
=======================================
Files 145 145
Lines 10984 10984
=======================================
Hits 10779 10779
Misses 205 205
Continue to review full report at Codecov.
|
There was a problem hiding this comment.
Thanks @josh146! This thing is getting massive!
I made a few cosmetic changes manually (you should check my commits to see if you like them).
Otherwise I have two main suggestions for polishing before release:
- Let's put the Hessians much lower. This was already advertised last release (though lower in the UI), and I want to give space for the newer changes like shots control, transforms, and new operations
- I got annoyed at how inconsistent the usage of pycon was (whether to have
>>>or not (sometimes even in the same example!). I tried to change a few, but eventually I gave up. We could either enforce a uniform style here or just not bother. What do you think?
.github/CHANGELOG.md
Outdated
| In addition, if computing the expectation value of a Hamiltonian, weighted random sampling can be | ||
| used to further distribute the shot budget across the local terms from which the Hamiltonian is | ||
| constructed. | ||
| * A new transform `qml.transforms.invisible` has been added. |
There was a problem hiding this comment.
I'm not keen on a few things here:
- the name
invisible - having it exposed as a public function
- encouraging its use at the QNode level
- having
.queue()in the example - having it as a "new feature"
There was a problem hiding this comment.
Should I remove it from the changelog? I originally was not going to include it here (just have it in 'improvements'), but I had a couple of discussions with some power users that indicated it might be useful to make this more discoverable.
There was a problem hiding this comment.
removing it from this PR only solves 2 of the 5 pain points above 😆
Co-authored-by: Nathan Killoran <co9olguy@users.noreply.github.com>
done 👍
tbh it doesn't really bother me; as long as the intention of the example is clear, I am happy. Enforcing consistency here I feel would result in a reduction in readability. So I am in favour of 'just not bother'. I had a look at other projects, and they also mix-and-match as needed, c.f. NumPy v1.20 release notes, and PyTorch. Personally, my preference is typically the examples that use a mix of both styles:
I find it nicely separates the code example vs. showing expected output. |
.github/CHANGELOG.md
Outdated
| quantum operation processing from being recorded by the QNode: | ||
|
|
||
| ```pycon | ||
| >>> @qml.transforms.invisible |
There was a problem hiding this comment.
I'm afraid I'm not a fan of this implementation. It just feels like the wrong UI compared to everything else we've already established. Would recommend revisiting.
No description provided.