用win7-64 bit-account编译python与＆＃39; hypot＆＃39;错误

Question

首先，我对Python是全新的，所以对以下内容的任何帮助都将非常感激。 我在Python中执行一些测试编码（使用pycharm作为前端），后者又调用了keras中的机器学习函数。

我创建了一个虚拟环境，基本解释器并设法运行第一个测试代码。但是，当我运行第二个代码（附在下面）时，我收到以下错误（也在下面附上）。

这个错误很可能与所谓的＆＃39; hypot＆＃39;如本问题所述的声明：compile some code with boost.python by mingw in win7-64bit。提到的答案提到：＆＃34;要解决这个问题，请在包含boost / python标题之前使用#include。＆＃34;

如果这也是我必须做的事情，请你帮我解决如何在包含boost / python标题之前包含cmath的问题＆＃39 ;? 任何想法都会非常感激。

CODE

{
 ...
  },
  "dependencies": {
    ...
    "css-loader": "0.28.4",
    "es6-shim": "0.35.3",
    "event-source-polyfill": "0.0.9",
    "expose-loader": "0.7.3",
    "extract-text-webpack-plugin": "2.1.2",
    "file-loader": "0.11.2",
    "html-loader": "0.4.5",
    "isomorphic-fetch": "2.2.1",
    "jquery": "3.2.1",
    "json-loader": "0.5.4",
    "preboot": "4.5.2",
    "raw-loader": "0.5.1",
    "reflect-metadata": "0.1.10",
    "rxjs": "5.5.5",
    "style-loader": "0.18.2",
    "to-string-loader": "1.1.5",
    "typescript": "2.4.1",
    "url-loader": "0.5.9",
    "webpack": "2.5.1",
    "webpack-hot-middleware": "2.18.2",
    "webpack-merge": "4.1.0",
    "zone.js": "0.8.12"
  },

ERROR：

# Segmentation demo

#import PIL

#from keras.utils import plot_model
# from keras.utils import plot_model
import keras.backend as K
import matplotlib.pyplot as plt
import numpy as np
from PIL import ImageDraw, Image
from keras.layers import Input, Conv2D
from keras.models import Model
# import PIL


# Generate random data
def generate_data(num):
    x, y = [], []
    for i in range(num):
        im = Image.new('RGB', (50, 50), (0, 0, 0))
        im_mask = Image.new('RGB', (50, 50), (0, 0, 0))

        draw = ImageDraw.Draw(im)
        draw_mask = ImageDraw.Draw(im_mask)

        num_shapes = 7
        edge_max = 8
        max_coord = 45

        # rectangles
        for i in range(np.random.randint(2, num_shapes)):
            edge = np.random.randint(2, edge_max)
            tl = (np.random.randint(max_coord), np.random.randint(max_coord))
            tr = (tl[0], tl[1] + edge)
            bl = (tl[0] + edge, tl[1])
            br = (tl[0] + edge, tl[1] + edge)
            draw.polygon([tl, bl, br, tr], fill="red")
            draw_mask.polygon([tl, bl, br, tr], fill="white")

        # lines
        for i in range(np.random.randint(2, num_shapes)):
            edge = np.random.randint(2, edge_max)
            a = (np.random.randint(max_coord - edge), np.random.randint(max_coord - edge))
            b = (a[0] + np.random.randint(2 * edge), a[1] + np.random.randint(2 * edge))
            draw.line((a, b), fill="red")

        # ellipses
        # for i in range(np.random.randint(2, num_shapes)):
        #   edge = np.random.randint(2, edge_max)
        #   tl = (np.random.randint(max_coord), np.random.randint(max_coord))
        #   br = (tl[0] + edge, tl[1] + edge)
        #   draw.ellipse([tl, br], fill="red")

        # pentagons
        for i in range(np.random.randint(2, num_shapes)):
            edge = np.random.randint(2, edge_max)
            tl2 = (np.random.randint(max_coord), np.random.randint(max_coord))
            tr = (tl2[0], tl2[1] + np.random.randint(edge))
            bl1 = (tl2[0] + np.random.randint(edge), tl2[1] + np.random.randint(edge))
            bl2 = (bl1[0] + np.random.randint(edge), bl1[1] + np.random.randint(edge))
            br = (bl2[0] + np.random.randint(edge), bl2[1] + np.random.randint(edge))
            draw.polygon([tl2, bl1, bl2, br, tr], fill="red")

        x.append(np.reshape(im.getdata(), (50, 50, 3)).T)
        y.append(np.reshape(im_mask.getdata(), (50, 50, 3)).T[0:1, :, :])

    x = np.array(x)
    y = np.array(y)
    return x, y


def plot_ims(model):
    x, y = generate_data(10)
    y[y == 255] = 1
    ypred = model.predict(x)

    plt.figure()
    for i in range(0, 6):
        plt.subplot(2, 6, i * 2 + 1)
        plt.imshow(x[i * 2 - 1, 0])
        plt.xticks([])
        plt.yticks([])
        plt.subplot(2, 6, i * 2 + 2)
        plt.imshow(ypred[i * 2 - 1, 0], cmap=plt.cm.gray)
        plt.xticks([])
        plt.yticks([])


def dsc(y_true, y_pred):
    y_true_f = K.flatten(y_true)
    y_pred_f = K.flatten(y_pred)
    intersection = K.sum(y_true_f * y_pred_f)
    return (2. * intersection + 1.) / (K.sum(y_true_f) + K.sum(y_pred_f) + 1.)


def dsc_loss(y_true, y_pred):
    return 1. - dsc(y_true, y_pred)


# Define Model

inp = Input(shape=(3, 50, 50))
l = Conv2D(8, 3, activation='relu', padding='same')(inp)
l = Conv2D(8, 3, activation='relu', padding='same')(l)
l = Conv2D(8, 3, activation='relu', padding='same')(l)
l = Conv2D(8, 3, activation='relu', padding='same')(l)
out = Conv2D(1, 1, activation='sigmoid', padding='same')(l)
model = Model(inputs=inp, outputs=out)
model.summary()
#plot_model(model, to_file='shapes_model.png', show_shapes=True, show_layer_names=False)
model.compile(loss=dsc_loss, optimizer='adam', metrics=['mse', dsc])
# model.compile(loss='mse', optimizer='adam', metrics=['mse', dsc])

#plot_ims(model)
plt.savefig('shapes_before_training.png')
plt.show()

eps = 5
x, y = generate_data(2000)
y[y == 255] = 1

h = model.fit(x, y, validation_split=0.1, batch_size=32, epochs=eps)

#plot_ims(model)
plt.savefig('shapes_after_training.png')
plt.show()

#Plot training / validation accuracy

plt.figure()
plt.title('Shapes Segmentation')
plt.plot(range(1, eps + 1), h.history['dsc'], label='Training', color='k')
plt.plot(range(1, eps + 1), h.history['val_dsc'], label='Validation', color='r')
plt.xticks(range(1, eps + 1))
plt.xlabel('Epochs')
plt.ylabel('DICE')
plt.legend(loc='best')
plt.savefig('shapes_training.png')
plt.show()