Contents

ALLCools.integration.harmony

ALLCools.integration.harmony

Module Contents

logger[source]
ch[source]
formatter[source]
leiden_centroids(pcs, n_pcs=None, n_neighbors=15, resolution=1.5)[source]

Instead of kmeans, using leiden algorithm to find centroids

run_harmony(data_mat: numpy.ndarray, meta_data: pandas.DataFrame, vars_use, theta=None, lamb=None, sigma=0.1, nclust=None, tau=0, block_size=0.05, max_iter_harmony=10, max_iter_kmeans=20, epsilon_cluster=1e-05, epsilon_harmony=0.0001, verbose=True, random_state=0, init_method='kmeans', n_pcs=None, n_neighbors=15, resolution=1.5, leiden_input='origin')[source]

Run Harmony.

class Harmony(Z, Phi, Phi_moe, Pr_b, sigma, theta, max_iter_harmony, max_iter_kmeans, epsilon_kmeans, epsilon_harmony, K, block_size, lamb, verbose, init_method, n_pcs, n_neighbors, resolution, leiden_input)[source]

Bases: object

result(self)[source]
allocate_buffers(self)[source]
init_cluster(self)[source]
compute_objective(self)[source]
harmonize(self, iter_harmony=10, verbose=True)[source]
cluster(self)[source]
update_R(self)[source]
check_convergence(self, i_type)[source]
safe_entropy(x: numpy.array)[source]
moe_correct_ridge(Z_orig, Z_cos, Z_corr, R, W, K, Phi_Rk, Phi_moe, lamb)[source]

previous

ALLCools.integration.confusion

next

ALLCools.mcds