t-SNE可视化将训练样本单独显示

最新推荐文章于 2024-02-04 08:00:00 发布

LUQC638

最新推荐文章于 2024-02-04 08:00:00 发布

阅读量502

点赞数 1

分类专栏： pytorch 文章标签： python 深度学习机器学习

本文链接：https://blog.csdn.net/weixin_61445075/article/details/125594786

版权

pytorch 专栏收录该内容

13 篇文章 0 订阅

订阅专栏

def plot_embedding(data, label, train_index,title):
    x_min, x_max = np.min(data, 0), np.max(data, 0)
    data = (data - x_min) / (x_max - x_min)

    fig = plt.figure()
    ax = plt.subplot(111)
    for i in range(data.shape[0]):
        if i not in train_index:
            plt.text(data[i, 0], data[i, 1], str(label[i]),
                     color=plt.cm.Set1(label[i] / 10.),
                     fontdict={'weight': 'bold', 'size': 9})
        else:
            plt.text(data[i, 0], data[i, 1], str(label[i]),
                     color=plt.cm.Set3(label[i] / 10.),
                     fontdict={'weight': 'bold', 'size': 9})
    plt.xticks([])
    plt.yticks([])
    plt.title(title)
    plt.show()

if __name__ == '__main__':

    # pdist = torch.nn.PairwiseDistance()
    data = torch.tensor([[0.1, 0.1,0.3],
                      [0.5, 0.6,0.5],
                         [0.6,0.7,0.8]])
    data = data.numpy()
    label = torch.tensor([0,0,1])
    label = label.numpy()
    train_index = [1]
    n_samples = data.shape[0]
    n_features = data.shape[1]
    print(data,label,n_samples,n_features)

    tsne = TSNE(n_components=3, init='pca', random_state=0)
    result = tsne.fit_transform(data)
    # fig = plot_embedding(result[:,0:2], label,'t-SNE embedding of the data')
    plot_embedding(result, label, train_index,'t-SNE embedding of the data')

颜色和字体的设计可自行调整