如何修改np.histogram直方图的Y轴数据为dots/m²？

你之前的错误在于误用了np.histogram——这个函数是用来统计数据分布的，你已经通过分组逻辑得到了每个区间的点数num_sizes，并且通过calculate_dots_per_m2算出了转换后的num_of_dots_per_m2，直接用这个数组作为柱状图的高度即可，不需要修改np.histogram的输入参数。

修改后的完整代码 #

import numpy as np
from matplotlib import pyplot as plt

color_dict = {
    "RED": (0, 0, 255),
    "GREEN": (0, 225, 0),
    "BLUE": (255, 0, 0),
    "ORANGE": (0, 165, 255),
    "MAGENTA": (255, 0, 255)
    }

color_texts = list(color_dict.keys())
colors = list(color_dict.values())
all_areas = [0.01, 0.1, 1, 10, 100]

def calculate_dots_per_m2(num_sizes):
    scale_factor = 25
    num_of_dots_per_m2 = []
    for i, num_of_defects in enumerate(num_sizes):
        num_of_dots_per_m2.append(num_of_defects*scale_factor)
    total_dots_per_m2 = sum(num_of_dots_per_m2)
    return num_of_dots_per_m2, total_dots_per_m2

def make_histogram():
    dots = [[] for i in range(len(all_areas))]
    all_dot_areas = [0.05, 1.4, 1.5, 1.9, 3.7, 11.5, 29.1, 99.0, 111.1, 4123.1]

    for dot_area in all_dot_areas:
        for i, a in enumerate(all_areas[:-1]):
            if all_areas[i] < dot_area < all_areas[i + 1]:
                dots[i].append(dot_area)
        if dot_area > all_areas[-1]:
            dots[-1].append(dot_area)

    num_sizes = [len(subls) for subls in dots]
    num_of_dots_per_m2, total_dots_per_m2 = calculate_dots_per_m2(num_sizes)

    bins = all_areas + [1e9]
    # 这里不需要修改np.histogram，因为hist和num_sizes结果一致，我们直接用转换后的num_of_dots_per_m2画图
    hist, bin_edges = np.histogram(all_dot_areas, bins)
    fig, ax = plt.subplots()
    fig.set_size_inches(16, 9)
    # 修改Y轴标签为dots/m²
    ylabel = "Number of dots per m^2"
    plt.ylabel(ylabel)

    ticklabels = []
    for i, a in enumerate(all_areas[:-1]):
        # 可选：把标签里的数量也改成dots/m²的数值
        size_x_label = "Dots sized {}-{} mm^2: {}".format(all_areas[i], all_areas[i + 1], num_of_dots_per_m2[i])
        ticklabels.append(size_x_label)
    # 同样修改最后一个标签的数值
    lastlabel = "Dots with size over {} mm^2: {}".format(all_areas[-1], num_of_dots_per_m2[-1])
    ticklabels.append(lastlabel)

    colours = [c.lower() for c in color_texts]
    # 把hist替换为num_of_dots_per_m2，用转换后的数值作为柱子高度
    ax.bar(range(len(num_of_dots_per_m2)), num_of_dots_per_m2, width=0.75, align="center", tick_label=ticklabels, color=colours, ec="black")
    plt.show()

make_histogram()

关键修改点 #

• 柱状图的高度数据从hist替换为num_of_dots_per_m2，直接使用转换后的数值
• 更新Y轴标签为Number of dots per m^2
• （可选）更新X轴标签中的数量显示，让标签和图表数据保持一致

这样修改后，所有柱状图的高度会变为原来的25倍，X轴的区间划分和描述保持不变，完全符合你的需求。

内容的提问来源于stack exchange，提问作者programer