当前位置：文江博客话题详情

热图 Cbar 调整颜色渐变

发布于 2025-01-11 13:15:53 字数 970 浏览 1 评论 0原文

我尝试调整我的 cbar 颜色，但遇到了困难：

当我使用：Center=None时，我有以下栏：

当我使用 Center=0 时 ，我有以下栏：

但这不是我所排除的：我的目标是从零开始为红色，并且值小于 0 时变得越来越红。

所以条形图应该看起来像这样的东西即：

这样，它会更具可读性：红色负数 - 绿色正数

（红色越多，负数越多 - 绿色越多，正数越多）

下面是我的代码行：

ax = sns.heatmap(result, linewidth=0.1,cmap="RdYlGn",
cbar_kws=dict(ticks= [result.min()/2,.0,result.max()/2]),
center=None,vmin=result.min(),vmax=result.max(),robust=True,fmt="f")

似乎 vmin 和 vmax 无法正常工作

原文

I try to adjust my cbar color but struggling with it :

When I use : Center=None I have the following bar :

And When I use Center=0 , I have the following bar :

But it is not what I except : My goal is to have red starting at zero and being more and more red more the value is less than 0.

So the bar should look Like stuff like that :

So like this it will be more readable : Red Negative - Green Positive

(More its Red More its negative - More its Green more it is positive)

Below is my line of code :

ax = sns.heatmap(result, linewidth=0.1,cmap="RdYlGn",
cbar_kws=dict(ticks= [result.min()/2,.0,result.max()/2]),
center=None,vmin=result.min(),vmax=result.max(),robust=True,fmt="f")

It seems that vmin and vmax does not work as excepted

分享到QQ

分享到微博

如果你对这篇内容有疑问，欢迎到本站社区发帖提问参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。

发布评论

需要登录才能够评论，你可以免费注册一个本站的账号。

无法回应 2025-01-18 13:15:53

您似乎想要一个以不同方式压缩两个部分的 TwoSlopeNorm 。（默认使用对称着色。）

from matplotlib import pyplot as plt
from matplotlib.colors import TwoSlopeNorm
import seaborn as sns

norm = TwoSlopeNorm(vcenter=0)
result = np.random.uniform(-0.01, 0.1, (10, 10))
ax = sns.heatmap(result, linewidth=0.1, cmap="RdYlGn", norm=norm,
                 cbar_kws=dict(ticks=[result.min() / 2, .0, result.max() / 2]),
                 robust=True, fmt="f")
plt.show()

You seem to want a TwoSlopeNorm which compresses both parts differently. (The default uses symmetric coloring.)

from matplotlib import pyplot as plt
from matplotlib.colors import TwoSlopeNorm
import seaborn as sns

norm = TwoSlopeNorm(vcenter=0)
result = np.random.uniform(-0.01, 0.1, (10, 10))
ax = sns.heatmap(result, linewidth=0.1, cmap="RdYlGn", norm=norm,
                 cbar_kws=dict(ticks=[result.min() / 2, .0, result.max() / 2]),
                 robust=True, fmt="f")
plt.show()