1

我的目标是在 PCA 缩减图旁边绘制一个轮廓图。我的想法是，为了便于理解，我希望在两个图表上使用相同的颜色。现在，我明白了：

在此处输入图像描述

我面临的问题是，在第一个情节中，我将每个剪影一个接一个地绘制，并且我有一个颜色列表，而对于第二个情节，一切都是同时绘制的。

所以我不知道如何从一种模式切换到另一种模式。

这是代码，它应该是一个工作示例。

def silhouette_PCA(data, model, n):
    reduced_data = sklearn.decomposition.PCA(n_components=2).fit_transform(data)
    model.fit(reduced_data)

    fig, (ax1, ax2) = plt.subplots(1, 2)
    fig.set_size_inches(18, 7)

    sample_silhouette_values = sklearn.metrics.silhouette_samples(reduced_data, model.fit_predict(reduced_data)  )

    y_lower = 10
    for i in range(n):
        ith_cluster_silhouette_values = sample_silhouette_values[ model.fit_predict(reduced_data) == i]
        ith_cluster_silhouette_values.sort()
        size_cluster_i = ith_cluster_silhouette_values.shape[0]
        y_upper = y_lower + size_cluster_i
        ############################### first color
        color = plt.cm.nipy_spectral(float(i) / n)
        ax1.fill_betweenx(np.arange(y_lower, y_upper),
                          0, ith_cluster_silhouette_values,
                          facecolor=color, edgecolor=color, alpha=0.7)

        y_lower = y_upper + 10 

#########################################################################################
    h = .02

    x_min, x_max = reduced_data[:, 0].min() - 1, reduced_data[:, 0].max() + 1
    y_min, y_max = reduced_data[:, 1].min() - 1, reduced_data[:, 1].max() + 1
    xx, yy = np.meshgrid(np.arange(x_min, x_max, h), np.arange(y_min, y_max, h))

    Z = model.predict(np.c_[xx.ravel(), yy.ravel()])

    Z = Z.reshape(xx.shape)
    ax2.imshow(Z, interpolation='nearest',
              extent=(xx.min(), xx.max(), yy.min(), yy.max()),
               ############################# here the 2nd Color
              cmap=plt.cm.Paired,
              aspect='auto', origin='lower')
    ax2.plot(reduced_data[:, 0], reduced_data[:, 1], 'k.', markersize=2)


    plt.show()


model = sklearn.cluster.KMeans(n_clusters = 3)
data = feat_matrix
silhouette_PCA(data,model,3)

谢谢你。

4

1 回答 1

2

我已经做到了，它工作正常：

def silhouette_PCA(data, model, n):
    reduced_data = sklearn.decomposition.PCA(n_components=2).fit_transform(data)
    model.fit(reduced_data)

    fig, (ax1, ax2) = plt.subplots(1, 2)
    fig.set_size_inches(18, 7)

    sample_silhouette_values = sklearn.metrics.silhouette_samples(reduced_data, model.fit_predict(reduced_data)  )

    y_lower = 10
    for i in range(n):
        ith_cluster_silhouette_values = sample_silhouette_values[ model.fit_predict(reduced_data) == i]
        ith_cluster_silhouette_values.sort()
        size_cluster_i = ith_cluster_silhouette_values.shape[0]
        y_upper = y_lower + size_cluster_i
        ############################### first color
        color = plt.cm.nipy_spectral(float(i) / n)
        ax1.fill_betweenx(np.arange(y_lower, y_upper),
                          0, ith_cluster_silhouette_values,
                          facecolor=color, edgecolor=color, alpha=0.7)

        y_lower = y_upper + 10 

    list = []
    for i in range(n):
      list = np.append(list , plt.cm.nipy_spectral(float(i) / n)    )
    list = np.reshape(list,  (n,4) )
    cmap = mpl.colors.ListedColormap(list)
    bounds= range(n)
    norm = mpl.colors.BoundaryNorm(bounds, cmap.N)
#########################################################################################
    h = .02

    x_min, x_max = reduced_data[:, 0].min() - 1, reduced_data[:, 0].max() + 1
    y_min, y_max = reduced_data[:, 1].min() - 1, reduced_data[:, 1].max() + 1
    xx, yy = np.meshgrid(np.arange(x_min, x_max, h), np.arange(y_min, y_max, h))

    Z = model.predict(np.c_[xx.ravel(), yy.ravel()])
    Z = Z.reshape(xx.shape)
    ax2.imshow(Z, interpolation='nearest',
              extent=(xx.min(), xx.max(), yy.min(), yy.max()),
               ############################# here the 2nd Color
              cmap= cmap, #plt.cm.Paired,
              aspect='auto', origin='lower')
    ax2.plot(reduced_data[:, 0], reduced_data[:, 1], 'k.', markersize=2)


    plt.show()


model = sklearn.cluster.KMeans(n_clusters = 7)
data = feat_matrix
silhouette_PCA(data,model,7)

在此处输入图像描述

在此处输入图像描述

于 2019-12-23T22:17:04.857 回答