[Return-types Fix #4] Update Hamiltonian expand transform

[eddddddy]

Context:
Part of the ongoing return types project

Description of the Change:
Update the hamiltonian_expand transform to work with the new return types.

[github-actions]

Hello. You may have forgotten to update the changelog!
Please edit doc/releases/changelog-dev.md with:

• A one-to-two sentence description of the change. You may include a small working example for new features.
• A link back to this PR.
• Your name (or GitHub username) in the contributors section.

[codecov]

Codecov Report

Merging #3232 (761ae5a) into master (5f332df) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master    #3232   +/-   ##
=======================================
  Coverage   99.70%   99.70%           
=======================================
  Files         277      277           
  Lines       24542    24544    +2     
=======================================
+ Hits        24469    24471    +2     
  Misses         73       73           

Impacted Files	Coverage Δ
pennylane/transforms/hamiltonian_expand.py	100.00% <100.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[rmoyard]

Thanks for the fix @eddddddy, it is looking good! 💯 Do you know if we should worry about multiple measurement and expectations values of different hamiltonian or different measurements? (see the example below). I think it should be separated in two different cases but I am not totally sure.

@qnode(dev, diff_method=diff_method, mode=mode, max_diff=max_diff)
def circuit(data, weights, coeffs):
   weights = weights.reshape(1, -1)
   qml.templates.AngleEmbedding(data, wires=[0, 1])
   qml.templates.BasicEntanglerLayers(weights, wires=[0, 1])
   H = qml.Hamiltonian(coeffs, obs)
   H.compute_grouping()
   H_2= qml.Hamiltonian(coeffs_2, obs_2)
   H_2.compute_grouping()
   return qml.expval(H), qml.expval(H2)

[eddddddy]

Hi @rmoyard, it seems that we don't have to worry about multiple Hamiltonians being measured:

pennylane/pennylane/_device.py

Lines 737 to 743 in cf195af

	try:
	circuits, hamiltonian_fn = qml.transforms.hamiltonian_expand(circuit, group=False)
	except ValueError as e:
	raise ValueError(
	"Can only return the expectation of a single Hamiltonian observable"
	) from e

Running your code, we get the following Exception:

Traceback (most recent call last):
  File "C:\Users\Edward\Documents\pennylane\pennylane\_device.py", line 738, in batch_transform
    circuits, hamiltonian_fn = qml.transforms.hamiltonian_expand(circuit, group=False)
  File "C:\Users\Edward\Documents\pennylane\pennylane\transforms\hamiltonian_expand.py", line 128, in hamiltonian_expand
    raise ValueError(
ValueError: Passed tape must end in `qml.expval(H)`, where H is of type `qml.Hamiltonian`

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "../scripts/return.py", line 159, in <module>
    main()
  File "../scripts/return.py", line 154, in main
    debug2()
  File "../scripts/return.py", line 146, in debug2
    circuit()
  File "C:\Users\Edward\Documents\pennylane\pennylane\qnode.py", line 622, in __call__
    res = qml.execute(
  File "C:\Users\Edward\Documents\pennylane\pennylane\interfaces\execution.py", line 575, in execute
    return _execute_new(
  File "C:\Users\Edward\Documents\pennylane\pennylane\interfaces\execution.py", line 347, in _execute_new
    tapes, batch_fn = qml.transforms.map_batch_transform(device.batch_transform, tapes)
  File "C:\Users\Edward\Documents\pennylane\pennylane\transforms\batch_transform.py", line 471, in map_batch_transform
    new_tapes, fn = transform(t)
  File "C:\Users\Edward\Documents\pennylane\pennylane\_device.py", line 741, in batch_transform
    raise ValueError(
ValueError: Can only return the expectation of a single Hamiltonian observable

[rmoyard]

Thanks @eddddddy 💯

[antalszava]

@eddddddy a nice fix! 👏 Just 2 questions.

[antalszava]

Looks good! 👍 Thanks for the answers.