Modernize code base (#38)

* Fix installation on non-UTF-8 platforms

Installation was broken on Windows,
because of non_ASCII characters in the README.

* Add Github actions for testing/coverage

Make sure that python-louain is installed for coverage.

* Remove travis

* Support Python versions 3.8-3.11

* Fix pytest deprecation warnings

We got a bunch of warnings in CI like this:

   tests/test_linkpred.py::TestLinkpred::test_predict_all
    /home/runner/work/linkpred/linkpred/.tox/py/lib/python3.11/site-packages/_pytest/fixtures.py:917: PytestRemovedIn8Warning: Support for nose tests is deprecated and will be removed in a future release.
    tests/test_linkpred.py::TestLinkpred::test_predict_all is using nose-specific method: `teardown(self)`
    To remove this warning, rename it to `teardown_method(self)`
    See docs: https://docs.pytest.org/en/stable/deprecations.html#support-for-tests-written-for-nose

* Add ruff and fix a bunch of ruff warnings

* Fix various deprecation warnings

This introduced a bug during development (sparse matrix vs sparse
array).
Since scipy 1.10, the power operator ** performs elementwise
multiplications, instead of matrix multiplication. In this change, we
implement a small recursive function to reinstate the old behaviour,
such that the Katz predictor can rely on it again.

Fixes #37

* Bump dependency versions

.github/workflows/coverage.yml

@@ -0,0 +1,34 @@
+name: Calculate test coverage
+
+on: [push]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+    - name: Check out
+      uses: actions/checkout@v2
+    - name: Set up Python
+      uses: actions/setup-python@v2
+      with:
+        python-version: "3.10"
+    - uses: actions/cache@v2
+      name: Configure pip caching
+      with:
+        path: ~/.cache/pip
+        key: ${{ runner.os }}-pip-${{ hashFiles('**/setup.py') }}
+        restore-keys: |
+          ${{ runner.os }}-pip-
+    - name: Install Python dependencies
+      run: |
+        pip install --upgrade pip
+        pip install -e .[community]
+        pip install pytest pytest-cov
+    - name: Run tests
+      run: |-
+        pytest --cov=linkpred --cov-report xml:coverage.xml --cov-report term
+    - name: Upload coverage report
+      uses: codecov/codecov-action@v1
+      with:
+        token: ${{ secrets.CODECOV_TOKEN }}
+        file: coverage.xml

.github/workflows/tox.yml

@@ -0,0 +1,31 @@
+name: Test with tox
+
+on: [push]
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    strategy:
+      max-parallel: 5
+      matrix:
+        python-version: ["3.8", "3.9", "3.10", "3.11"]
+
+    steps:
+    - uses: actions/checkout@v2
+    - name: Set up Python ${{ matrix.python-version }}
+      uses: actions/setup-python@v2
+      with:
+        python-version: ${{ matrix.python-version }}
+    - uses: actions/cache@v2
+      name: Configure pip caching
+      with:
+        path: ~/.cache/pip
+        key: ${{ runner.os }}-pip-${{ hashFiles('**/setup.py') }}
+        restore-keys: |
+          ${{ runner.os }}-pip-
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install tox tox-gh-actions
+    - name: Test with tox
+      run: tox

linkpred/cli.py

@@ -4,6 +4,8 @@
 import logging
 import sys
 
+import yaml
+
 from .exceptions import LinkPredError
 from .linkpred import LinkPred
 from .predictors import all_predictors
@@ -26,17 +28,12 @@ def load_profile(fname):
     """Load the JSON or YAML profile with the given filename"""
     try:
         with open(fname) as f:
-            if fname.endswith(".yaml") or fname.endswith(".yml"):
-                import yaml
-
+            if fname.endswith((".yaml", ".yml")):
                 return yaml.safe_load(f)
-            else:
-                return json.load(f)
-    except Exception as e:
-        raise LinkPredError(
-            "Encountered error while loading profile '%s'. "
-            "Error message: '%s'" % (fname, e)
-        )
+            return json.load(f)
+    except Exception as err:
+        msg = f"Encountered error while loading profile '{fname}'. "
+        raise LinkPredError(msg) from err
 
 
 def get_config(args=None):
@@ -142,7 +139,7 @@ def handle_arguments(args=None):
         metavar="PREDICTOR",
     )
 
-    all_help = "Predict all links, " "including ones present in the training network"
+    all_help = "Predict all links, including ones present in the training network"
     parser.add_argument(
         "-a",
         "--all",

linkpred/evaluation/listeners.py

@@ -1,11 +1,11 @@
 import copy
 import logging
-import smokesignal
-
 from time import localtime, strftime
 
-from .static import EvaluationSheet
+import smokesignal
+
 from ..util import interpolate
+from .static import EvaluationSheet
 
 log = logging.getLogger(__name__)
 
@@ -28,7 +28,7 @@ def _timestamped_filename(basename, ext="txt"):
     return basename + strftime("_%Y-%m-%d_%H.%M.", localtime()) + ext
 
 
-class Listener(object):
+class Listener:
     def __init__(self):
         smokesignal.on("dataset_finished", self.on_dataset_finished)
         smokesignal.on("run_finished", self.on_run_finished)
@@ -45,7 +45,7 @@ def __init__(self, **kwargs):
         smokesignal.on("prediction_finished", self.on_prediction_finished)
         self.params = kwargs
 
-        super(EvaluatingListener, self).__init__()
+        super().__init__()
 
     def on_prediction_finished(self, scoresheet, dataset, predictor):
         evaluation = EvaluationSheet(scoresheet, **self.params)
@@ -60,21 +60,21 @@ def on_prediction_finished(self, scoresheet, dataset, predictor):
 class CachePredictionListener(Listener):
     def __init__(self):
         smokesignal.on("prediction_finished", self.on_prediction_finished)
-        super(CachePredictionListener, self).__init__()
+        super().__init__()
         self.encoding = "utf-8"
 
     def on_prediction_finished(self, scoresheet, dataset, predictor):
-        self.fname = _timestamped_filename("%s-%s-predictions" % (dataset, predictor))
+        self.fname = _timestamped_filename(f"{dataset}-{predictor}-predictions")
         scoresheet.to_file(self.fname)
 
 
 class CacheEvaluationListener(Listener):
     def __init__(self):
         smokesignal.on("evaluation_finished", self.on_evaluation_finished)
-        super(CacheEvaluationListener, self).__init__()
+        super().__init__()
 
     def on_evaluation_finished(self, evaluation, dataset, predictor):
-        self.fname = _timestamped_filename("%s-%s-predictions" % (dataset, predictor))
+        self.fname = _timestamped_filename(f"{dataset}-{predictor}-predictions")
         evaluation.to_file(self.fname)
 
 
@@ -84,16 +84,16 @@ def __init__(self, name, beta=1):
         self.fname = _timestamped_filename("%s-Fmax" % name)
 
         smokesignal.on("evaluation_finished", self.on_evaluation_finished)
-        super(FMaxListener, self).__init__()
+        super().__init__()
 
     def on_evaluation_finished(self, evaluation, dataset, predictor):
         fmax = evaluation.f_score(self.beta).max()
 
-        status = "%s\t%s\t%.4f\n" % (dataset, predictor, fmax)
+        status = f"{dataset}\t{predictor}\t{fmax:.4f}\n"
 
         with open(self.fname, "a") as f:
             f.write(status)
-        print(status)
+        log.info("Evaluation finished: %s", status)
 
 
 class PrecisionAtKListener(Listener):
@@ -102,15 +102,15 @@ def __init__(self, name, k=10):
         self.fname = _timestamped_filename("%s-precision-at-%d" % (name, self.k))
 
         smokesignal.on("evaluation_finished", self.on_evaluation_finished)
-        super(PrecisionAtKListener, self).__init__()
+        super().__init__()
 
     def on_evaluation_finished(self, evaluation, dataset, predictor):
         precision = evaluation.precision()[self.k]
 
-        status = "%s\t%s\t%.4f\n" % (dataset, predictor, precision)
+        status = f"{dataset}\t{predictor}\t{precision:.4f}\n"
         with open(self.fname, "a") as f:
             f.write(status)
-        print(status)
+        log.info("Evaluation finished: %s", status)
 
 
 GENERIC_CHART_LOOKS = [
@@ -157,14 +157,14 @@ def __init__(self, name, xlabel="", ylabel="", filetype="pdf", chart_looks=None)
         self._y = []
 
         smokesignal.on("evaluation_finished", self.on_evaluation_finished)
-        super(Plotter, self).__init__()
+        super().__init__()
 
     def add_line(self, predictor=""):
         ax = self.fig.axes[0]
         ax.plot(self._x, self._y, self.chart_look(), label=predictor)
 
         log.debug(
-            "Added line with %d points: " "start = (%.2f, %.2f), end = (%.2f, %.2f)",
+            "Added line with %d points: start = (%.2f, %.2f), end = (%.2f, %.2f)",
             len(self._x),
             self._x[0],
             self._y[0],
@@ -190,16 +190,16 @@ def on_run_finished(self):
 
         # Save to file
         self.fname = _timestamped_filename(
-            "%s-%s" % (self.name, self._charttype), self.filetype
+            f"{self.name}-{self._charttype}", self.filetype
         )
         self.fig.savefig(self.fname)
 
 
 class RecallPrecisionPlotter(Plotter):
     def __init__(
-        self, name, xlabel="Recall", ylabel="Precision", interpolation=True, **kwargs
+        self, name, xlabel="Recall", ylabel="Precision", *, interpolation=True, **kwargs
     ):
-        super(RecallPrecisionPlotter, self).__init__(name, xlabel, ylabel, **kwargs)
+        super().__init__(name, xlabel, ylabel, **kwargs)
         self._charttype = "recall-precision"
         self.interpolation = interpolation
 
@@ -215,7 +215,7 @@ def setup_coords(self, evaluation):
 
 class FScorePlotter(Plotter):
     def __init__(self, name, xlabel="#", ylabel="F-score", beta=1, **kwargs):
-        super(FScorePlotter, self).__init__(name, xlabel, ylabel, **kwargs)
+        super().__init__(name, xlabel, ylabel, **kwargs)
         self._charttype = "F-Score"
         self.beta = beta
 
@@ -228,7 +228,7 @@ class ROCPlotter(Plotter):
     def __init__(
         self, name, xlabel="False pos. rate", ylabel="True pos. rate", **kwargs
     ):
-        super(ROCPlotter, self).__init__(name, xlabel, ylabel, **kwargs)
+        super().__init__(name, xlabel, ylabel, **kwargs)
         self._charttype = "ROC"
 
     def setup_coords(self, evaluation):
@@ -238,7 +238,7 @@ def setup_coords(self, evaluation):
 
 class MarkednessPlotter(Plotter):
     def __init__(self, name, xlabel="Miss", ylabel="Precision", **kwargs):
-        super(MarkednessPlotter, self).__init__(name, xlabel, ylabel, **kwargs)
+        super().__init__(name, xlabel, ylabel, **kwargs)
         self._charttype = "Markedness"
         self._legend_props["loc"] = "upper left"
 

linkpred/evaluation/scoresheet.py

@@ -1,7 +1,7 @@
 import logging
-import networkx as nx
-
 from collections import defaultdict
+
+import networkx as nx
 from networkx.readwrite.pajek import make_qstr
 
 log = logging.getLogger(__name__)
@@ -34,7 +34,7 @@ def __init__(self, data=None):
     def __setitem__(self, key, val):
         dict.__setitem__(self, key, float(val))
 
-    def process_data(self, data, *args, **kwargs):
+    def process_data(self, data):
         """Can be overridden by child classes"""
         return data
 
@@ -75,7 +75,7 @@ def from_record(line, delimiter="\t"):
     @staticmethod
     def to_record(key, value, delimiter="\t"):
         key, value = map(make_qstr, (key, value))
-        return u"{}{}{}\n".format(key, delimiter, value)
+        return f"{key}{delimiter}{value}\n"
 
     @classmethod
     def from_file(cls, fname, delimiter="\t", encoding="utf-8"):
@@ -94,7 +94,7 @@ def to_file(self, fname, delimiter="\t", encoding="utf-8"):
                 fh.write(self.to_record(key, score, delimiter).encode(encoding))
 
 
-class Pair(object):
+class Pair:
     """An unsorted pair of things.
 
     We could probably also use frozenset for this, but a Pair class opens
@@ -122,9 +122,10 @@ def __init__(self, *args):
         elif len(args) == 2:
             a, b = args
         else:
-            raise TypeError("__init__() takes 1 or 2 arguments in addition to self")
+            msg = "__init__() takes 1 or 2 arguments in addition to self"
+            raise TypeError(msg)
         # For link prediction, a and b are two different nodes
-        assert a != b, "Predicted link (%s, %s) is a self-loop!" % (a, b)
+        assert a != b, f"Predicted link ({a}, {b}) is a self-loop!"
         self.elements = self._sorted_tuple((a, b))
 
     @staticmethod
@@ -216,4 +217,4 @@ def from_record(line, delimiter="\t"):
     def to_record(key, value, delimiter="\t"):
         u, v = key
         u, v, score = map(make_qstr, (u, v, value))
-        return u"{0}{3}{1}{3}{2}\n".format(u, v, score, delimiter)
+        return f"{u}{delimiter}{v}{delimiter}{score}\n"