python - 在 Python 中优化和改进绘图

Question

我正在尝试制作描述为即将举行的比赛注册的竞争对手的情节。目前，我有一个制作绘图的工作代码，但我想让绘图有线条而不是点，并且要更改 y 限制。这听起来很简单，但我一直遇到我没想到的错误。这可能是因为我的代码效率不高。

我当前的数据保存在两个不同的数组中，一个具有 349 个单独的分区，按 4 个分区特征：颜色、年龄、性别和体重分区，如下所示：

[u'BLUE', u'Master 1', u'Male', u'Heavy'],
[u'BLUE', u'Master 1', u'Male', u'Super-Heavy'],
[u'BLUE', u'Master 1', u'Male', u'Ultra-Heavy'],
[u'BLUE', u'Master 1', u'Female', u'Rooster'],
[u'BLUE', u'Master 1', u'Female', u'Light-Feather'],
[u'BLUE', u'Master 1', u'Female', u'Feather'],
[u'BLUE', u'Master 1', u'Female', u'Light'],
[u'BLUE', u'Master 1', u'Female', u'Middle'],
[u'BLUE', u'Master 1', u'Female', u'Medium-Heavy'],

然后我有一个 349 长度的数组，其中包含每个部门的竞争对手数量。

我最初的目标是制作一个面具，以便挑选出相同的颜色、年龄和性别划分，然后按重量绘制它们。我还想制作一个图，将相同颜色和性别的所有不同年龄组相加，并按体重计算该腰带颜色和性别的总注册人。

该代码不喜欢针对字符串（用于权重划分）进行绘图，因此我还制作了包含不同权重划分的字典作为 x 轴：

male_adult_divisions = {"Rooster":127.0, "Light-Feather":141.5, "Feather":154.5, "Light":168.0, "Middle":181.5, "Medium-Heavy":195.0, "Heavy":208.0, "Super-Heavy":222.0, "Ultra-Heavy":235.0}
male_juvenile1_divisions = {"Rooster":107.0, "Light-Feather":118.0, "Feather":129.0, "Light":141.0, "Middle":152.0, "Medium-Heavy":163.0, "Heavy":175.0, "Super-Heavy":186.0, "Ultra-Heavy":197.0}
male_juvenile2_divisions = {"Rooster":118.0, "Light-Feather":129.0, "Feather":141.5, "Light":152.5, "Middle":163.5, "Medium-Heavy":175.0, "Heavy":186.0, "Super-Heavy":197.0, "Ultra-Heavy":208.0}
female_adult_divisions = {"Rooster":107.0, "Light-Feather":118.0, "Feather":129.0, "Light":141.0, "Middle":152.0, "Medium-Heavy":163.0, "Heavy":175.0, "Super-Heavy":186.0}
female_juvenile_divisions = {"Rooster":98.0, "Light-Feather":106.5, "Feather":116.0, "Light":125.0, "Middle":133.5, "Medium-Heavy":143.5, "Heavy":152.0, "Super-Heavy":160.5}

这是我到目前为止工作的代码：

for gender in gender_divisions:
 gender_index = numpy.where(numpy.transpose(divisions)[2] == gender)

 for age in age_divisions:
    age_index = numpy.where(numpy.transpose(divisions)[1] == age)

    if (age == "Juvenile 1 "or age == "Juvenile 2 ") and gender == "Female":
        weight_limits = female_juvenile_divisions

    elif age == "Juvenile 1 "and gender == "Male":
        weight_limits = male_juvenile1_divisions

    elif age == "Juvenile 2 "and gender == "Male":
        weight_limits = male_juvenile2_divisions

    elif gender == "Female":
        weight_limits = female_adult_divisions

    else:
        weight_limits = male_adult_divisions

    mask = numpy.intersect1d(gender_index,age_index)
    mask = numpy.intersect1d(mask, exclude_absolute)

    if not mask.size:
        break

    for i,color in enumerate(belt_divisions):
        boolean_mask = numpy.zeros(len(numpy.transpose(divisions)[0]))

        if color == "WHITE":
            white_mask = numpy.intersect1d(mask, numpy.where(color == numpy.transpose(divisions)[0]))
            for x in white_mask:
                plt.plot(weight_limits[numpy.transpose(divisions[x])[3]],competitors[x], marker = "o", c="white")

        elif color == "BLUE":
            blue_mask = numpy.intersect1d(mask, numpy.where(color == numpy.transpose(divisions)[0]))
            for x in blue_mask:
                plt.plot(weight_limits[numpy.transpose(divisions[x])[3]],competitors[x], marker = "o", c="blue")

        elif color == "PURPLE":
            purple_mask = numpy.intersect1d(mask, numpy.where(color == numpy.transpose(divisions)[0]))
            for x in purple_mask:
                plt.plot(weight_limits[numpy.transpose(divisions[x])[3]],competitors[x], marker = "o", c="violet")

        elif color == "BROWN":
            brown_mask = numpy.intersect1d(mask, numpy.where(color == numpy.transpose(divisions)[0]))
            for x in brown_mask:
                plt.plot(weight_limits[numpy.transpose(divisions[x])[3]],competitors[x], marker = "o", c="brown")

        else:
            black_mask = numpy.intersect1d(mask, numpy.where(color == numpy.transpose(divisions)[0]))
            for x in black_mask:
                plt.plot(weight_limits[numpy.transpose(divisions[x])[3]],competitors[x], marker = "o", c="black")


        plt.title("Competitors in %s %s Divisions"%(gender,age))
        plt.xlabel("Weight Category")
        plt.ylabel("Number of Competitors")
        plt.ylim(0,)

    plt.show()

if age != "Juvenile 1" and age != "Juvenile 2":
    if gender == "Female":
        weight_limits = female_adult_divisions
        mask = numpy.intersect1d(gender_index, exclude_absolute)

        for i,color in enumerate(belt_divisions):
            boolean_mask = numpy.zeros(len(numpy.transpose(divisions)[0]))

            if color == "WHITE":
                white_mask = numpy.intersect1d(mask, numpy.where(color == numpy.transpose(divisions)[0]))
                for x in white_mask:
                    plt.plot(weight_limits[numpy.transpose(divisions[x])[3]],competitors[x], marker = "o", c="white")

            elif color == "BLUE":
                blue_mask = numpy.intersect1d(mask, numpy.where(color == numpy.transpose(divisions)[0]))
                for x in blue_mask:
                    plt.plot(weight_limits[numpy.transpose(divisions[x])[3]],competitors[x], marker = "o", c="blue")

            elif color == "PURPLE":
                purple_mask = numpy.intersect1d(mask, numpy.where(color == numpy.transpose(divisions)[0]))
                for x in purple_mask:
                    plt.plot(weight_limits[numpy.transpose(divisions[x])[3]],competitors[x], marker = "o", c="violet")

            elif color == "BROWN":
                brown_mask = numpy.intersect1d(mask, numpy.where(color == numpy.transpose(divisions)[0]))
                for x in brown_mask:
                    plt.plot(weight_limits[numpy.transpose(divisions[x])[3]],competitors[x], marker = "o", c="brown")

            else:
                black_mask = numpy.intersect1d(mask, numpy.where(color == numpy.transpose(divisions)[0]))
                for x in black_mask:
                    plt.plot(weight_limits[numpy.transpose(divisions[x])[3]],competitors[x], marker = "o", c="black")


        plt.title("Competitors in All %s Divisions"%(gender))
        plt.xlabel("Weight Category")
        plt.ylabel("Number of Competitors")
        plt.ylim(0,)

    plt.show()

当我最初试图写这个时，我试图创建一个索引数组来挑选我想要的分区，但由于某种原因，我无法让它工作。

例如，如果我有black_mask = array([343, 344, 345, 346, 347, 348])，那么我尝试做divisions[black_mask]，我得到：

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-17-c1f99913b04a> in <module>()
----> 1 divisions[black_mask]

TypeError: only integer arrays with one element can be converted to an index

所以我想也许我需要制作布尔数组以将其用作掩码：

boolean_mask = numpy.zeros(len(numpy.transpose(divisions)[0]))
for x in black_mask:
    boolean_mask[x] = 1

但我得到了同样的错误，这就是为什么我最终选择了上面效率低下的代码。

然后我想改变情节的 y 限制。我通过创建一个名为 max_ 的变量并将其设置为零来做到这一点。每次代码输入皮带颜色的 if 语句之一时，我都会输入if competitors[x]> max_: max_ = competitors[x]，但由于某种原因，这使我的所有图都变为空白（即使我没有将其放入 y 限制中，我认为）。

我也想把它做成一个折线图，但由于必须逐点放置，我无法做到。

所以目前这些是我要出去的地块类型：

这就是我想要效仿的：

标签： pythonmatplotliboptimizationplot