add option to disable edge rescaling to provide edge weights as integers, closes #1

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "lang-main"
-version = "0.1.2dev1"
+version = "0.1.2"
 description = "Several tools to analyse TOM's data with strong focus on language processing"
 authors = [
     {name = "d-opt GmbH, resp. Florian Förster", email = "f.foerster@d-opt.com"},
@@ -132,7 +132,7 @@ directory = "reports/coverage"
 
 
 [tool.bumpversion]
-current_version = "0.1.2dev1"
+current_version = "0.1.2"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

src/lang_main/analysis/graphs.py

@@ -467,6 +467,27 @@ def pipe_rescale_graph_edge_weights(
     return graph.rescale_edge_weights()
 
 
+def pipe_graph_split(
+    graph: TokenGraph,
+) -> tuple[TokenGraph, Graph]:
+    """splits TokenGraph instance into itself and its undirected version
+
+    Parameters
+    ----------
+    graph : TokenGraph
+        token graph pushed through pipeline
+
+    Returns
+    -------
+    tuple[TokenGraph, Graph]
+        token graph (directed) and undirected version with no changes made
+    """
+    graph = graph.copy()
+    graph_undir = graph.to_undirected(inplace=False)
+
+    return graph, graph_undir
+
+
 def normalise_array_linear(
     array: npt.NDArray[np.float32],
 ) -> npt.NDArray[np.float32]:

src/lang_main/pipelines/predefined.py

@@ -164,11 +164,15 @@ def build_tk_graph_post_pipe() -> Pipeline:
 def build_tk_graph_rescaling_pipe(
     save_result: bool,
     exit_point: EntryPoints,
+    enable_rescaling: bool = True,
 ) -> Pipeline:
     pipe_graph_rescaling = Pipeline(name='Graph_Rescaling', working_dir=SAVE_PATH_FOLDER)
-    pipe_graph_rescaling.add(
+    if enable_rescaling:
-        graphs.pipe_rescale_graph_edge_weights,
+        pipe_graph_rescaling.add(
-    )
+            graphs.pipe_rescale_graph_edge_weights,
+        )
+    else:
+        pipe_graph_rescaling.add(graphs.pipe_graph_split)
     pipe_graph_rescaling.add(
         graphs.pipe_add_graph_metrics,
         save_result=save_result,

tests/analysis/test_graphs.py

@@ -326,6 +326,15 @@ def test_pipe_rescale_graph_edge_weights(tk_graph):
     assert rescaled_undir[1][2]['weight'] == pytest.approx(1.0)
 
 
+def test_pipe_graph_split(tk_graph):
+    graph_copy, graph_undir = graphs.pipe_graph_split(tk_graph)
+    assert len(graph_copy.nodes) == len(tk_graph.nodes)
+    assert len(graph_copy.edges) == len(tk_graph.edges)
+    assert len(graph_copy.nodes) == len(graph_undir.nodes)
+    assert len(graph_undir.nodes) == len(tk_graph.nodes)
+    assert len(graph_undir.edges) != len(tk_graph.edges)
+
+
 @pytest.mark.parametrize('import_graph', ['graph', 'tk_graph'])
 def test_rescale_edge_weights(import_graph, request):
     test_graph = request.getfixturevalue(import_graph)

tests/analysis/test_preprocessing.py

@@ -29,13 +29,6 @@ def test_remove_na(raw_data_path, raw_data_date_cols):
     assert len(data) == 998
 
 
-# def test_string_cleansing():
-#     string = 'Ölleckage   durch\nundichten \t Ölsumpf,, aber Dichtung intakt??!!!'
-#     cleaned_string = shared.clean_string_slim(string)
-#     target_string = 'Ölleckage durch undichten Ölsumpf, aber Dichtung intakt!'
-#     assert cleaned_string == target_string
-
-
 def test_entry_wise_cleansing(raw_data_path, raw_data_date_cols):
     (data,) = ppc.load_raw_data(raw_data_path, raw_data_date_cols)
     (data,) = ppc.remove_duplicates(data)

tests/pipelines/test_predefined.py

@@ -28,14 +28,26 @@ def test_build_tk_graph_post_pipe():
     assert len(pipe.actions) == 3
 
 
-def test_build_tk_graph_rescaling_pipe():
+def test_build_tk_graph_rescaling_pipe_with_edge_rescaling():
     pipe = pre.build_tk_graph_rescaling_pipe(
-        save_result=False, exit_point=EntryPoints.TK_GRAPH_ANALYSIS_RESCALED
+        save_result=False,
+        exit_point=EntryPoints.TK_GRAPH_ANALYSIS_RESCALED,
+        enable_rescaling=True,
     )
     assert pipe.name == 'Graph_Rescaling'
     assert len(pipe.actions) == 2
 
 
+def test_build_tk_graph_rescaling_pipe_without_edge_rescaling():
+    pipe = pre.build_tk_graph_rescaling_pipe(
+        save_result=False,
+        exit_point=EntryPoints.TK_GRAPH_ANALYSIS_RESCALED,
+        enable_rescaling=False,
+    )
+    assert pipe.name == 'Graph_Rescaling'
+    assert len(pipe.actions) == 1
+
+
 @pytest.mark.parametrize('with_subgraphs', [True, False])
 def test_build_tk_graph_render_pipe(with_subgraphs):
     pipe = pre.build_tk_graph_render_pipe(with_subgraphs=with_subgraphs)