种类	模型	参数	fit	partial_fit	transfrom	fit_transfrom	inverse_transfrom	n_jobs	方法/字段	官网
decomposition	PCA	n_components : int, float, None or string whiten : bool, optional (default False)     可以保证输出值的方差为1 svd_solver : {‘auto’, ‘full’, ‘arpack’, ‘randomized’}	有	书上说有 但实际无	有	有	有	无	get_covariance() Compute data covariance with the generative model. get_params([deep]) Get parameters for this estimator. get_precision() Compute data precision matrix with the generative model. inverse_transform(X) Transform data back to its original space.逆操作升维 score(X[, y]) Return the average log-likelihood of all samples. score_samples(X) Return the log-likelihood of each sample. set_params(**params) Set the parameters of this estimator. components_ explained_variance_ explained_variance_ratio_	http://scikit-learn.org/stable/modules/generated/sklearn.decomposition.PCA.html
	KernelPCA	n_components : int, default=None kernel : “linear” | “poly” | “rbf” | “sigmoid” | “cosine” | “precomputed” gamma : float, default=1/n_features,针对rbf, poly and sigmoid degree : int, default=3 . Degree针对 poly coef0 : float, default=1. poly and sigmoid中的自由项 kernel_params ：自定义Kernel中的参数 alpha : int, default=1.0	有	无	有	有	有			http://scikit-learn.org/stable/modules/generated/sklearn.decomposition.KernelPCA.html
manifold	MDS	n_components : int, optional, default: 2，要压缩到多少维 metric : boolean, optional, default: True，     True用metric度量(适合连续数据)，False用nometric度量（适合分类变量） n_init : int, optional, default: 4，重复运行几次 dissimilarity : ‘euclidean’(default) | ‘precomputed’, optional     ‘euclidean’输入的是原始数据     ‘precomputed’输入的是距离矩阵	有		无	有	无		stress_#压力值，可以用来计算应当降为多少维 embedding_#mds没有transform，这个是降维后的数据 stress_#压力矩阵，可以理解为误差	http://scikit-learn.org/stable/modules/generated/sklearn.manifold.MDS.html
	Isomap	n_neighbors : integer：number of neighbors n_components : 要压缩到多少维 eigen_solver : [‘auto’|’arpack’|’dense’] path_method : string [‘auto’|’FW’|’D’] neighbors_algorithm : string [‘auto’|’brute’|’kd_tree’|’ball_tree’]	有	无	无	有	无		embedding_ kernel_pca_	http://scikit-learn.org/stable/modules/generated/sklearn.manifold.Isomap.html
	LLE	n_neighbors : integer n_components : integer reg : float正则化系数 eigen_solver : string, {‘auto’, ‘arpack’, ‘dense’}							embedding_vectors_ reconstruction_error_#误差	http://scikit-learn.org/stable/modules/generated/sklearn.manifold.LocallyLinearEmbedding.html
neural_network	BernoulliRBM	n_components : int, 要压缩到多少维 learning_rate batch_size : int, optional n_iter : int, optional verbose : int, optional. The verbosity level. The default, zero, means silent mode. random_state	有	有	有	有	无		intercept_hidden_ : array-like, shape (n_components,) Biases of the hidden units. intercept_visible_ : array-like, shape (n_features,) Biases of the visible units. components_ : array-like, shape (n_components, n_features) Weight matrix, where n_features in the number of visible units and n_components is the number of hidden units.	https://scikit-learn.org/stable/modules/generated/sklearn.neural_network.BernoulliRBM.html