Hacktoberfest 2026:维护者为十月标记出来的 issue,仍然开放、适合新手。 浏览 Hacktoberfest issue

axis_scale='unit' centres on the mean, so skewed data sit off-centre

未关闭
#287 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

还没有人认领这个 Issue。

评估

难度
4/5
预计耗时
3-5 天
新手友好度
22/100
Issue 类型
缺陷
描述清晰度
基本清楚
活跃度
活跃
技术栈
matplotlib, plotly, python

调研方向

The centring logic for axis_scale='unit' is the starting point, and the issue traces it to the mean rather than the midrange. Run tests/test_gh141_marker_line_smoothing.py and tests/test_plot_audit_b1.py::TestStaticLineDataFidelity first, since these five tests encode the current placement and will fail under the change. The issue defers this to 1.2 and it touches every 2-D and 3-D figure, so confirm scope with maintainers before editing.

由索引模型根据 Issue 内容生成。

描述

enhancement

With the default axis_scale='unit', each axis is centred on the data's mean, not its midrange. Skewed data therefore sit off-centre in the plot box.

Measured during the 1.1 release review (candidate c433b8f0):

  • hyp.plot(...) 2-D on a random-walk fixture (feature tour PLOT-basic-*): the trajectory fills the top two-thirds of the box and the bottom third is empty. Matplotlib and plotly behave the same.
  • weights_sample: every axis mean maps to about +0.13. Per axis, 72–76% of points sit above the centre, and 41% fall in the upper octant.
  • A lognormal array: means map to about −0.85, and 0.7% of points per axis sit above the centre. The data fill the lower corner.

Centring each axis on its midrange (min + max) / 2 would fill the box. It changes how every 2-D and 3-D figure looks. On a patched copy, 5 tests that encode the current placement fail: 3 in tests/test_gh141_marker_line_smoothing.py and 2 in tests/test_plot_audit_b1.py::TestStaticLineDataFidelity. Every gallery image, tutorial output and screenshot would need regenerating, so this is deferred from 1.1 to 1.2.

主要语言
Python
星标
1.9k
派生
164
平均合并
2 小时 46 分钟
30 天内合并 PR
2

环境准备

  • 没有 Dockerfile 或 Docker Compose 文件
  • 没有 Pull Request 模板
  • 阅读贡献指南

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

ContextLab/hypertools 的其他 Issue

查看 ContextLab/hypertools 的全部 Issue

相似的 Issue

更多 Python Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。