phylovelo.embedding

Classes

VelocityEmbedding

Functions

paired_correlation_rows(→ numpy.array)	Calculate paired correlation
_as_float_array(x)
velocity_embedding(sd[, target, n_neigh, chunk_size])	Project velocity into embedding

Module Contents

paired_correlation_rows(A: numpy.array, B: numpy.array) → numpy.array

Calculate paired correlation

Args:

A: numpy.array B: numpy.array

Return:

numpy.array

_as_float_array(x)

class VelocityEmbedding(count, xdr, v)

count

xdr

kNN = None

v

d

_d_mean

_d_centered

_d_norm

_d_nonzero

_d_nonzero_values

_neighbor_indices = None

neighs_log

rho(x)

get_neighbors(kNN)

_get_neighbor_indices()

_auto_chunk_size(n_neighs, n_genes, chunk_size=None)

_transition_probabilities(diff_vecs)

_project_neighbor_block(start, stop, neighs)

project_all(chunk_size=None, store_transition=False)

transit_mat1(i)

transit_mat(n_process=0)

project(i)

velocity_embedding(sd: scData, target: str = 'count', n_neigh: int = None, chunk_size: int = None)

Project velocity into embedding

Args:

sd:

scData

target:

count or x_normed

n_neigh:

kNN pooling. Default: Ncells//3

chunk_size:

Number of cells per vectorized block. Default estimates a memory-safe size.