Releases: KrishnaswamyLab/graphtools
Releases · KrishnaswamyLab/graphtools
v1.5.3
What's Changed
- Deploy to PyPi on tagged release by @scottgigante in #65
- Use
scipy.sparse.csgraph.shortest_path
by @scottgigante in #66
Full Changelog: v1.5.2...v1.5.3
graphtools v1.5.0
Changeset:
- added
graphtools.estimator
abstract class: allows inheritance of basic takes-data-builds-graph functionality.
graphtools v1.4.0
Changeset:
- added
knn_max
argument tokNNGraph
- increased test coverage
- restyled code with
black
graphtools v1.3.0
Changeset:
shortest_path
now works for all graphs withdata
andaffinity
distance metrics- added automatic
n_pca
detection (thanks @stanleyjs) - changed default
decay
to 40 - renamed
kernel_symm='theta'
tokernel_symm='mnn'
- bugfixes
graphtools v1.1.0
Changeset:
- Changed default value of
theta
forkernel_symm='theta'
to 1 (was 0.5) - Added
BaseGraph.to_pickle
andgraphtools.read_pickle
functionality for saving and storing graphs - Improved warning message for many duplicate entries
- various bug fixes
graphtools v1.0.0
Changeset:
knn
no longer includes self. Note that this change means that to reproduce a graph created withgraphtools v0.*
, you will need to reduceknn
by 1.- deprecated Python 2.7
- fixed some inefficient sparse matrix functions
graphtools v0.2.1
Changes:
- added anisotropic kernel (alpha parameter from Coifman and Lafon 2006)
- added
to_igraph
function - allowed
from_igraph
to return weighted kernel - documentation updates
graphtools v0.2.0
Changeset:
- significant changes to
MNNGraph
kernel creation: between batch connections are now downweighted relative to the sum of within-batch connections - added fixed bandwidth kernel creation with
bandwidth
parameter - exposed
clusters
attribute onLandmarkGraph
- added igraph->graphtools conversion function
- added graphtools->PyGSP conversion function
MNNGraph
parametergamma
renamed totheta
graphtools v0.1.10
Changes:
- Uses tasklogger for logging
- Allows duplicates in building the graph; connections evaluated at nan are set to 1 with a warning.
- Landmarking is now done on the symmetric diffusion affinity matrix for true spectral clustering
- LandmarkGraph now exposes the transition matrix from samples to landmarks.
graphtools v0.1.9
Graphtools 0.1.9.
Changes:
graphtools.base.Data
acceptsanndata.AnnData
inputgraphtools.base.Data.inverse_transform
accepts column indices to return only a subset of ambient dimensions- minor bugfixes and improvements to documentation