Analyse temporal network and hypergraphs efficiently.
Project description
Python bindings for Reticula
Installation
The library offers pre-compiled Wheels for manylinux2014
compatible systems.
That is, Linux systems with GNU C Library (glibc) version 2.17 and newer. The
library currently supports Python version 3.8, 3.9 and 3.10.
$ pip install reticula
Installing from source
Alternatively you can install the library from source:
Clone the library:
$ git clone https://github.com/arashbm/reticula-python.git
Build the Wheel:
$ cd reticula-python
$ pip install .
Note that compiling source requires an unbelievable amount (> 40GB) of RAM.
Basic examples
Generate a random static network and investigate:
import reticula as ret
state = ret.mersenne_twister(42) # create a pseudorandom number generator
g = ret.random_gnp_graph[ret.int64](n=100, p=0.02, random_state=state)
print(g) # => <undirected_network[int64] with 100 verts and 110 edges>
print(g.vertices()) # => [0, 1, 2, 3, .... 99]
print(g.edges())
# => [undirected_edge[int64](0, 16), undirected_edge[int64](0, 20),
# undirected_edge[int64](0, 31), undirected_edge[int64](0, 51),
# ...]
print(ret.connected_components(g))
# => [<component[int64] of 1 nodes: {9}>,
# <component[int64] of 1 node {33}>,
# ...]
lcc = max(ret.connected_components(g), key=len)
print(lcc) # => <component[int64] of 93 nodes: {99, 96, 95, 94, ...}>
g2 = ret.vertex_induced_subgraph(g, lcc)
print(g2) # => <undirected_network[int64] with 93 verts and 109 edges>
A more complete example of static network percolation analysis, running on
multiple threads, can be found in
examples/static_network_percolation/
Create a random fully-mixed temporal network and calculate simple (unconstrained) reachability from node 0 at time 0 to all nodes and times.
import reticula as ret
state = ret.mersenne_twister(42)
g = ret.random_fully_mixed_temporal_network[ret.int64](
size=100, rate=0.01, max_t=1024, random_state=state)
adj = ret.temporal_adjacency.simple[
ret.undirected_temporal_edge[ret.int64, ret.double]]()
cluster = ret.out_cluster(
temporal_network=g, temporal_adjacency=adj, vertex=0, time=0.0)
print(cluster)
# => <temporal_cluster[undirected_temporal_edge[int64, double],
# simple[undirected_temporal_edge[int64, double]]] with volume 100
# and lifetime (0 1.7976931348623157e+308]>
print(cluster.covers(vertex=12, time=100.0)) # => True
# Let's see all intervals where vert 15 is reachable from vert 0 at t=0.0:
print(list(cluster.interval_sets()[15]))
# => [(3.099055278145548, 1.7976931348623157e+308)]
Let's now try limited waiting-time (with dt = 5.0) reachability:
import reticula as ret
state = ret.mersenne_twister(42)
g = ret.random_fully_mixed_temporal_network[int64](
size=100, rate=0.01, max_t=1024, random_state=state)
adj = ret.temporal_adjacency.limited_waiting_time[
ret.undirected_temporal_edge[ret.int64, ret.double]](dt=5.0)
cluster = ret.out_cluster(
temporal_network=g, temporal_adjacency=adj, vertex=0, time=0.0)
print(cluster)
# => <temporal_cluster[undirected_temporal_edge[int64, double],
# limited_waiting_time[undirected_temporal_edge[int64, double]]] with
# volume 100 and lifetime (0 1028.9972186553928]>
print(cluster.covers(vertex=15, time=16.0)) # => True
print(list(cluster.interval_sets()[15]))
# => [(3.099055278145548, 200.17866501023616),
# (200.39858803326402, 337.96139372380003),
# ...
# (1017.5258263596586, 1028.9149586273347)]
# Total "human-hours" of reachability cluster
print(cluster.mass()) # => 101747.97444555275
# Survival time of the reachability cluster
print(cluster.lifetime()) # => (0.0, 1028.9972186553928)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distributions
Hashes for reticula-0.1-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 215aa5d7e2190e576d36d97cf7a0e3f747826d937be6e3c1013e0ced7251530d |
|
MD5 | 736963365c5d12e6f1dbfd3767c6db9a |
|
BLAKE2b-256 | fbde02a07d5da6b45601e205b37264211e0230a687b3ec681be87d4797b0e207 |
Hashes for reticula-0.1-pp38-pypy38_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 235e6bb1fd423188834184ef08a78831731f9b1aa59b143dbc89403edf6fc35c |
|
MD5 | 45f941a11c6470616bba9e99f1d534db |
|
BLAKE2b-256 | 4a694b894244d23e2c050271e6e2a13dfafef9046161f253b4b0aa036a52cd3e |
Hashes for reticula-0.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 4e0dbd89f9845259645587bd635fe74c76f5506d9b7a98e74704852afdd99b56 |
|
MD5 | 8c7462fd2b8bbfe9f7c70c19617532e4 |
|
BLAKE2b-256 | 0b7ea8843d6e0326397345aa4b1c808926b2c56ab54f3f0a94edc3dd311288c7 |
Hashes for reticula-0.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 442444471cf932fd85cfc8d1cb908501981334c2bc51afede60da59396cad9df |
|
MD5 | a5aca523c7916937c092054207d0066b |
|
BLAKE2b-256 | e21f42e144bdc78d4b59259dc3325877f87bb80b0329126af26031a1fc5ab8fa |
Hashes for reticula-0.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 3017584948da36a40d0d07bc3dd66b38531ae52920a776e8185b228da036fee7 |
|
MD5 | a5755e7ad23f3049e1196b9484c1b92b |
|
BLAKE2b-256 | e44ca4da2b1e6e8b0c62e7125b2507b6446bd7b83a76401d1feafcca57948dd9 |