Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

docs(benchmarks): headline figure mislabels its steps, model and Majorana axis

Open
  • Is something wrong?
#402 0 comments 1 reaction 2 assignees View on GitHub

Maintainers usually reply within 1 day

@robertodr is already working on this.

Since Oct 17, 2026.

  • #411 by @copilot-swe-agent — open

Assessment

Difficulty
3/5
Estimated time
1-2 days
Newbie friendliness
82/100
Issue type
Bug
Clarity
Clearly specified
Activity status
Active
Tech stack
markdown, python
Domain
data-visualization, documentation

Research direction

Start with packages/bench-third-party/plot_headline.py, then check the benchmark settings and model descriptions cited in the issue to confirm the plotted step counts and runtime meanings. Update the plot and docs/content/docs/benchmarks.mdx to correct the Pauli point alignment, labels, model wording, and Majorana time axis; check the related plot scripts for the same issues. Done when the figure and caption match the data and the docs describe the model accurately, without rerunning the benchmark.

Written by the indexing model from the issue text.

Description

bug documentation

🤖 AI text below 🤖

The headline figure in docs/content/docs/benchmarks.mdx (/benchmarks/headline_runtime.png, drawn by packages/bench-third-party/plot_headline.py) has four labelling problems. The data are fine. Only the axes, the point placement and the titles are wrong, so the fix is in the plot script and the docs text, with no rerun.

1. Pauli points are drawn one label early

pauli_prop/run_model.py drops step 0's runtime (warm-up), so runtime has 27 entries against 28 in step_range (results.json: len(step_range) == 28, every runtime series has 27). plot_headline.py pairs them from the start:

points = [(s, v) for s, v in zip(steps, runtime) if s >= PAULI_MIN_STEP]

So the runtime of the k-th propagation call lands on the label of call k−1. It should be zip(steps[1:], runtime).

2. The x labels count by 2, but each point is one Trotter step

step_range is range(0, 55, 2) (settings.json: step_size: 2). But every iteration of _run_steps (backends.py) and of run_model.jl applies step_circuit once, and that is one Trotter step (model.py: "One Trotter step as a Qiskit circuit"). The point labelled "Trotter step 54" is therefore after 28 Trotter steps (t = 1.4 at dt = 0.05, not 2.7). The docs text already says "28 Trotter points". The fix is either to label the x axis with the step count (1…28), or to make step_size mean what it says. The second changes the benchmark, so it would need a rerun.

3. The Pauli model is tilted-field, not transverse-field

With hz = 1.0 (settings.json), the model has a longitudinal field and is a tilted-field Ising model, as model.py's docstring says ("a Trotterized 2D tilted-field Ising model"). The panel title in plot_headline.py ("Pauli: 144-qubit 2D transverse-field Ising") and the text in benchmarks.mdx ("2D transverse-field Ising model") say transverse-field.

4. The Majorana panel plots cumulative time on a "time per step" axis

majorana_prop/monoprop_hubbard1d_benchmark.py stores runtime_seconds as cumulative seconds (cumulative_runtimes[step + 1] = cumulative_runtimes[step] + step_runtime). plot_headline.py plots those values directly under the y label "time per step [s]". The fix is to plot np.diff(runtime_seconds) (per layer), or to relabel the axis as cumulative time. The caption's "complete Majorana evolution" ratio uses the total, so it is unaffected.

Scope
  • packages/bench-third-party/plot_headline.py: items 1–4.
  • docs/content/docs/benchmarks.mdx: item 3's wording, and the figure caption if the axes change.
  • Worth checking whether pauli_prop/plot_results.py and majorana_prop/plot_results.py share items 1, 2 and 4.
Dominant language
C++
Stars
52
Forks
2
Avg merge
18h 4m
Merged PRs (30d)
28

Getting set up

Open in Codespaces

Starts the project's dev container in your browser, under your own GitHub account.

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from Algorithmiq/monoprop

All issues in Algorithmiq/monoprop

Similar issues

More C++ issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.