Introduce naive layout algorithm

Author: PhilipMathieu

notebooks/tests.ipynb

@@ -7,10 +7,21 @@
 from .core.metricssuite import MetricsSuite
 from .core.readabilitygraph import ReadabilityGraph
 
-import utils.helpers as helpers
-import utils.crosses_promotion as crosses_promotion
+# Import layout algorithms
+from .layout.layout_algorithms import naive_optimizer
+
+# Import utils
+from .utils import helpers
+from .utils import crosses_promotion
 
 # Import tests
-import tests
+from . import tests
 
-__all__ = ["MetricsSuite", "ReadabilityGraph", "helpers", "crosses_promotion", "tests"]
+__all__ = [
+    "MetricsSuite",
+    "ReadabilityGraph",
+    "naive_optimizer",
+    "helpers",
+    "crosses_promotion",
+    "tests",
+]

src/__init__.py

@@ -1 +1,49 @@
-# Module for different graph layout algorithms (to be implemented).
+import networkx as nx
+from ..core.metricssuite import MetricsSuite
+
+
+def naive_optimizer(M_input: MetricsSuite, inplace=False):
+    """Naive optimizer for the graph layout.
+
+    This optimizer generates layouts based on four NetworkX graph layout algorithms:
+    - spring_layout
+    - shell_layout
+    - circular_layout
+    - kamada_kawai_layout
+    Then chooses the layout that maximizes the MetricSuite score.
+
+    Parameters
+    ----------
+    M : MetricSuite
+        The MetricSuite object to be optimized.
+    inplace : bool, optional
+        Whether to modify the MetricSuite object in place or return a new one. Default is False.
+
+    Returns
+    -------
+    MetricSuite
+        The optimized MetricSuite object.
+    """
+    layouts = {
+        "spring": nx.spring_layout,
+        "shell": nx.shell_layout,
+        "circular": nx.circular_layout,
+        "kamada_kawai": nx.kamada_kawai_layout,
+    }
+
+    scores = {}
+    poss = {}
+    for layout_name, layout_func in layouts.items():
+        M = M_input.copy() if not inplace else M_input
+        poss[layout_name] = layout_func(M._graph)
+        M.apply_layout(poss[layout_name])
+        M.calculate_metrics()
+        scores[layout_name] = M.combine_metrics()
+
+    # Choose the best layout
+    best_layout = max(scores, key=scores.get)
+    M = M_input.copy() if not inplace else M_input
+    M.apply_layout(poss[best_layout])
+
+    print(f"Best layout: {best_layout}. Score: {scores[best_layout]}")
+    return M