c71dea1c929f08d290142f50679a06303eb5bd9c,Python/phate/mds.py,,embed_MDS,#,37

Before Change

    // MDS embeddings, each gives a different output.
    X_dist = squareform(pdist(X, distance_metric))

    if how == "classic":
        // classical MDS as defined in cmdscale
        // Y = cmdscale(X_dist)[0][:,:ndim]
        Y = cmdscale_fast(X_dist, ndim)
    elif how == "metric":
        // First compute CMDS
        Y_cmds = cmdscale_fast(X_dist, ndim)
        // Metric MDS from sklearn
        Y = MDS(n_components=ndim, metric=True, max_iter=3000, eps=1e-12,
                dissimilarity="precomputed", random_state=seed, n_jobs=n_jobs,
                n_init=1).fit_transform(X_dist, init=Y_cmds)
    elif how == "nonmetric":
        // First compute CMDS
        Y_cmds = cmdscale_fast(X_dist, ndim)
        // Then compute Metric MDS
        Y_mmds = MDS(n_components=ndim, metric=True, max_iter=3000, eps=1e-12,
                     dissimilarity="precomputed", random_state=seed,
                     n_jobs=n_jobs, n_init=1).fit_transform(X_dist,
                                                            init=Y_cmds)
        // Nonmetric MDS from sklearn using metric MDS as an initialization
        Y = MDS(n_components=ndim, metric=False, max_iter=3000, eps=1e-12,
                dissimilarity="precomputed", random_state=seed, n_jobs=n_jobs,
                n_init=1).fit_transform(X_dist, init=Y_mmds)
    else:
        raise ValueError("Allowable "how" values for MDS: "classic", "
                         ""metric", or "nonmetric". "
                         ""{}" was passed.".format(how))
    return Y

After Change

    // initialize all by CMDS
    Y = cmdscale_fast(X_dist, ndim)
    if how in ["metric", "nonmetric"]:
        log_debug("Performing metric MDS on "
                  "{} of shape {}...".format(type(X_dist),
                                             X_dist.shape))
        // Metric MDS from sklearn
        Y = MDS(n_components=ndim, metric=True, max_iter=3000, eps=1e-12,
                dissimilarity="precomputed", random_state=seed, n_jobs=n_jobs,
                n_init=1).fit_transform(X_dist, init=Y)

Instances