我需要可视化一组二维数据，数据如下：

[[ 0. , 234. , 568.33333333, 881.66666667],
 [ 0. , 234.66888889, 569.44555556, 865.03806017],
 [ 0. , 235.33627407, 570.06910753, 849.9592539 ],
 [ 0. , 235.9989178 , 570.2632452 , 836.28943309]]

我尝试用Plotly的等高线图来展示，但发现列方向的细微数值差异完全无法在图中体现出来，我的现有代码如下：

import numpy as np
from plotly.subplots import make_subplots
import plotly.graph_objects as go
from pathlib import Path
import os
import plotly.io as pio

Smax = 1000
T = 1
classical_2d_without_boundary = np.array([
    [ 0. , 234. , 568.33333333, 881.66666667],
    [ 0. , 234.66888889, 569.44555556, 865.03806017],
    [ 0. , 235.33627407, 570.06910753, 849.9592539 ],
    [ 0. , 235.9989178 , 570.2632452 , 836.28943309]
])

n_rows, n_cols = classical_2d_without_boundary.shape
x = np.linspace(0, Smax, n_cols)
y = np.linspace(T, 0, n_rows)
x, y = np.meshgrid(x,y)

fig = make_subplots(rows= 1, cols = 1)
quantum_contour = go.Contour(
    z=classical_2d_without_boundary,
    x=x[0, :], # assuming x is 2D after meshgrid
    y=y[:, 0], # assuming y is 2D after meshgrid
    colorscale='Inferno'
)
fig.add_trace(quantum_contour, row = 1, col = 1)

fig.update_layout(
    title_x=0.5,
    title_font = dict(size=40),
    xaxis_title="x-axis",
    yaxis_title="time (Year)",
    height=900,
    width=855, # Earlier it was 750
    xaxis_title_font = dict(size=36),
    yaxis_title_font = dict(size= 36),
    legend_title_font= dict(size=36),
    font= dict(size=30),
    paper_bgcolor ='white',
    margin=dict(l=20, r=20, t=350, b=20)
)
fig.update_xaxes(title_font=dict(size=24))
fig.update_yaxes(title_font=dict(size=24))

for annotation in fig['layout']['annotations']:
    annotation['font'].update(size= 40)
    annotation.update(y=1.02) # Increase the y value to move the title up

# display
fig.show()

针对你遇到的“细微数值差异无法呈现”的问题，可以从以下几个方向调整：

1. 强制调整等高线的层级精度 #

默认的等高线层级会自动合并小的数值变化，你可以手动设置contours参数，生成更细密的层级，同时显示等高线线条来强化差异：

quantum_contour = go.Contour(
    z=classical_2d_without_boundary,
    x=x[0, :],
    y=y[:, 0],
    colorscale='Inferno',
    # 自定义等高线层级，聚焦数据实际范围
    contours=dict(
        start=np.min(classical_2d_without_boundary[:,1:]),  # 跳过全0的第一列
        end=np.max(classical_2d_without_boundary),
        size=0.3,  # 层级间隔，值越小越能捕捉细微差异
        showlines=True,  # 显示等高线线条
        line=dict(width=0.5)
    ),
    # 优化颜色条刻度，方便对应数值
    colorbar=dict(
        tickmode='linear',
        dtick=1,
        title=dict(text='数值', font=dict(size=20))
    )
)

2. 改用热力图(Heatmap)展示细节 #

热力图对局部数值变化的敏感度远高于等高线图，更适合展示小规模数据的细微差异：

# 替换Contour为Heatmap
quantum_heatmap = go.Heatmap(
    z=classical_2d_without_boundary,
    x=x[0, :],
    y=y[:, 0],
    colorscale='Inferno',
    colorbar=dict(
        tickmode='linear',
        dtick=1,
        title=dict(text='数值', font=dict(size=20))
    ),
    # 开启 hover 提示，方便查看精确数值
    hoverongaps = False
)
fig.add_trace(quantum_heatmap, row=1, col=1)

3. 消除无效数据的干扰 #

你的数据第一列全为0，这个大跨度的低值会压缩有效数据的颜色刻度范围，可以单独对有效列做可视化，或者对数据做列归一化：

# 方案1：只可视化非0的列
valid_data = classical_2d_without_boundary[:,1:]
x_valid = x[0, 1:]
quantum_contour = go.Contour(
    z=valid_data,
    x=x_valid,
    y=y[:, 0],
    colorscale='Inferno',
    contours=dict(size=0.3)
)

# 方案2：对每一列做归一化，聚焦列内相对变化
normalized_data = (classical_2d_without_boundary - classical_2d_without_boundary.min(axis=0)) / (classical_2d_without_boundary.max(axis=0) - classical_2d_without_boundary.min(axis=0))

4. 插值补充数据点 #

如果数据点太少，细节会被掩盖，可以用插值生成更密集的数据，放大变化趋势：

from scipy.interpolate import interp2d

# 生成插值后的密集数据
f = interp2d(x[0,:], y[:,0], classical_2d_without_boundary, kind='cubic')
x_new = np.linspace(0, Smax, 100)  # 从4个点扩展到100个点
y_new = np.linspace(T, 0, 40)
z_new = f(x_new, y_new)

# 用插值后的数据绘图
quantum_contour = go.Contour(
    z=z_new,
    x=x_new,
    y=y_new,
    colorscale='Inferno',
    contours=dict(size=0.2)
)

内容来源于stack exchange