Add has_path, inter_community_edges, intra_community_edges (#22)

* Add `has_path`, `inter_community_edges, `intra_community_edges`

* Add katz_centrality

* eigenvector_centrality

* Add `normalize` and `is_converged` helper functions

* Add degree centralities

* HITS

* Rename `mark_nx_tests` to `on_start_tests`

* Add NoteMap and NodeSet to make vectors look like sets and dicts

* change `has_path` to use bidirectional BFS

Author: eriknw

.github/dependabot.yml

@@ -0,0 +1,6 @@
+version: 2
+updates:
+  - package-ecosystem: 'github-actions'
+    directory: '/'
+    schedule:
+      interval: 'weekly'

.github/workflows/test.yml

@@ -33,7 +33,7 @@ jobs:
           conda install -c conda-forge python-graphblas scipy pandas \
             pytest-cov pytest-randomly black flake8-comprehensions flake8-bugbear
             # matplotlib lxml pygraphviz pydot sympy  # Extra networkx deps we don't need yet
-          pip install git+https://github.com/jim22k/networkx.git@nx-sparse --no-deps
+          pip install git+https://github.com/mriduls/networkx.git@nx-sparse --no-deps
           pip install -e . --no-deps
       - name: Style checks
         run: |

.pre-commit-config.yaml

@@ -2,18 +2,21 @@
 #
 # Before first use: `pre-commit install`
 # To run: `pre-commit run --all-files`
+fail_fast: true
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.2.0
+    rev: v4.3.0
     hooks:
       - id: check-added-large-files
+      - id: check-ast
+      - id: check-toml
       - id: check-yaml
       - id: debug-statements
       - id: end-of-file-fixer
       - id: mixed-line-ending
       - id: trailing-whitespace
   - repo: https://github.com/myint/autoflake
-    rev: v1.4
+    rev: v1.7.7
     hooks:
       - id: autoflake
         args: [--in-place]
@@ -23,34 +26,42 @@ repos:
       - id: isort
         language_version: python3
   - repo: https://github.com/asottile/pyupgrade
-    rev: v2.32.0
+    rev: v3.1.0
     hooks:
       - id: pyupgrade
         args: [--py38-plus]
+  - repo: https://github.com/MarcoGorelli/auto-walrus
+    rev: v0.2.1
+    hooks:
+    - id: auto-walrus
   - repo: https://github.com/psf/black
-    rev: 22.3.0
+    rev: 22.10.0
     hooks:
       - id: black
         language_version: python3
         args: [--target-version=py38]
   - repo: https://github.com/PyCQA/flake8
-    rev: 4.0.1
+    rev: 5.0.4
     hooks:
       - id: flake8
         additional_dependencies: &flake8_dependencies
-        - flake8==4.0.1
-        - flake8-comprehensions==3.8.0
-        - flake8-bugbear==22.3.23
+        - flake8==5.0.4
+        - flake8-comprehensions==3.10.0
+        - flake8-bugbear==22.9.23
   - repo: https://github.com/asottile/yesqa
-    rev: v1.3.0
+    rev: v1.4.0
     hooks:
       - id: yesqa
         additional_dependencies: *flake8_dependencies
   - repo: https://github.com/codespell-project/codespell
-    rev: v2.1.0
+    rev: v2.2.2
     hooks:
       - id: codespell
         types_or: [python, rst, markdown]
         files: ^(graphblas_algorithms|docs)/
         # args: ["--ignore-words-list=coo,ba"]
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v4.3.0
+    hooks:
+      - id: no-commit-to-branch  # no commit directly to main
 # Maybe: black-jupyter, blacken-docs, blackdoc mypy, velin

graphblas_algorithms/_utils.py

@@ -1,6 +1,8 @@
 from . import exceptions
 from .boundary import *
+from .centrality import *
 from .cluster import *
+from .community import *
 from .core import *
 from .cuts import *
 from .dag import *
@@ -9,6 +11,7 @@
 from .link_analysis import *
 from .reciprocity import *
 from .regular import *
+from .shortest_paths import *
 from .simple_paths import *
 from .smetric import *
 from .structuralholes import *

graphblas_algorithms/algorithms/__init__.py

@@ -0,0 +1,29 @@
+from graphblas import binary, monoid, unary
+
+
+def normalize(x, how):
+    how = how.lower()
+    if how == "l1":
+        denom = x.reduce().get(0)
+    elif how == "l2":
+        denom = (x @ x).get(0) ** 0.5
+    elif how == "linf":
+        denom = x.reduce(monoid.max).get(0)
+    else:
+        raise ValueError(f"Unknown normalization method: {how}")
+    try:
+        x *= 1.0 / denom
+    except ZeroDivisionError:  # pragma: no cover
+        pass
+    return x
+
+
+def is_converged(xprev, x, tol):
+    """Check convergence, L1 norm: err = sum(abs(xprev - x)); err < N * tol
+
+    This modifies `xprev`.
+    """
+    xprev << binary.minus(xprev | x)
+    xprev << unary.abs(xprev)
+    err = xprev.reduce().get(0)
+    return err < xprev.size * tol

graphblas_algorithms/algorithms/_helpers.py

@@ -25,5 +25,5 @@ def node_boundary(G, nbunch1, *, mask=None):
         mask = ~nbunch1.S
     else:
         mask = mask & (~nbunch1.S)
-    bdy = any_pair(G._A.T @ nbunch1).new(mask=mask, name="boundary")
+    bdy = any_pair(nbunch1 @ G._A).new(mask=mask, name="boundary")
     return bdy

graphblas_algorithms/algorithms/boundary.py

@@ -0,0 +1,3 @@
+from .degree_alg import *
+from .eigenvector import *
+from .katz import *

graphblas_algorithms/algorithms/centrality/__init__.py

@@ -0,0 +1,32 @@
+from graphblas import Vector
+
+__all__ = ["degree_centrality", "in_degree_centrality", "out_degree_centrality"]
+
+
+def _degree_centrality(G, degrees, name):
+    N = len(G)
+    rv = Vector(float, size=N, name=name)
+    if N <= 1:
+        rv << 1
+    else:
+        s = 1 / (N - 1)
+        rv << s * degrees
+    return rv
+
+
+def degree_centrality(G, *, name="degree_centrality"):
+    if G.is_directed():
+        degrees = G.get_property("total_degrees+")
+    else:
+        degrees = G.get_property("degrees+")
+    return _degree_centrality(G, degrees, name)
+
+
+def in_degree_centrality(G, *, name="in_degree_centrality"):
+    degrees = G.get_property("column_degrees+")
+    return _degree_centrality(G, degrees, name)
+
+
+def out_degree_centrality(G, *, name="out_degree_centrality"):
+    degrees = G.get_property("row_degrees+")
+    return _degree_centrality(G, degrees, name)

graphblas_algorithms/algorithms/centrality/degree_alg.py

@@ -0,0 +1,37 @@
+from graphblas import Vector
+
+from graphblas_algorithms.algorithms._helpers import is_converged, normalize
+from graphblas_algorithms.algorithms.exceptions import (
+    ConvergenceFailure,
+    GraphBlasAlgorithmException,
+    PointlessConcept,
+)
+
+__all__ = ["eigenvector_centrality"]
+
+
+def eigenvector_centrality(G, max_iter=100, tol=1.0e-6, nstart=None, name="eigenvector_centrality"):
+    N = len(G)
+    if N == 0:
+        raise PointlessConcept("cannot compute centrality for the null graph")
+    x = Vector(float, N, name="x")
+    if nstart is None:
+        x << 1.0 / N
+    else:
+        x << nstart
+        denom = x.reduce().get(0)  # why not use L2 norm?
+        if denom == 0:
+            raise GraphBlasAlgorithmException("initial vector cannot have all zero values")
+        x *= 1.0 / denom
+
+    # Power iteration: make up to max_iter iterations
+    A = G._A
+    xprev = Vector(float, N, name="x_prev")
+    for _ in range(max_iter):
+        xprev << x
+        x += x @ A
+        normalize(x, "L2")
+        if is_converged(xprev, x, tol):  # sum(abs(xprev - x)) < N * tol
+            x.name = name
+            return x
+    raise ConvergenceFailure(max_iter)