Refactor artifacts detection + add saturation detection

conftest.py

@@ -8,6 +8,21 @@ def create_cache_folder(tmp_path_factory):
     return cache_folder
 
 
+@pytest.fixture(scope="module")
+def debug_plots(request):
+    """Return True if debug plots should be shown."""
+    return request.config.getoption("--debug-plots")
+
+
+def pytest_addoption(parser):
+    parser.addoption(
+        "--debug-plots",
+        action="store_true",
+        default=False,
+        help="Enable debug plots during tests",
+    )
+
+
 def pytest_collection_modifyitems(config, items):
     """
     This function marks (in the pytest sense) the tests according to their name and file_path location

doc/api.rst

@@ -213,6 +213,9 @@ spikeinterface.preprocessing
     .. autofunction:: detect_bad_channels
     .. autofunction:: detect_and_interpolate_bad_channels
     .. autofunction:: detect_and_remove_bad_channels
+    .. autofunction:: detect_artifact_periods
+    .. autofunction:: detect_artifact_periods_by_envelope
+    .. autofunction:: detect_saturation_periods
     .. autofunction:: directional_derivative
     .. autofunction:: filter
     .. autofunction:: gaussian_filter

src/spikeinterface/core/node_pipeline.py

@@ -495,16 +495,17 @@ def find_parents_of_type(list_of_parents, parent_type):
     return parents
 
 
-def check_graph(nodes):
+def check_graph(nodes, check_for_peak_source=True):
     """
     Check that node list is orderd in a good (parents are before children)
     """
 
-    node0 = nodes[0]
-    if not isinstance(node0, PeakSource):
-        raise ValueError(
-            "Peak pipeline graph must have as first element a PeakSource (PeakDetector or PeakRetriever or SpikeRetriever"
-        )
+    if check_for_peak_source:
+        node0 = nodes[0]
+        if not isinstance(node0, PeakSource):
+            raise ValueError(
+                "Peak pipeline graph must have as first element a PeakSource (PeakDetector or PeakRetriever or SpikeRetriever"
+            )
 
     for i, node in enumerate(nodes):
         assert isinstance(node, PipelineNode), f"Node {node} is not an instance of PipelineNode"
@@ -532,6 +533,7 @@ def run_node_pipeline(
     verbose=False,
     skip_after_n_peaks=None,
     recording_slices=None,
+    check_for_peak_source=True,
 ):
     """
     Machinery to compute in parallel operations on peaks and traces.
@@ -587,6 +589,8 @@ def run_node_pipeline(
         Optionaly give a list of slices to run the pipeline only on some chunks of the recording.
         It must be a list of (segment_index, frame_start, frame_stop).
         If None (default), the function iterates over the entire duration of the recording.
+    check_for_peak_source : bool, default True
+        Whether to check that the first node is a PeakSource (PeakDetector or PeakRetriever or
 
     Returns
     -------
@@ -595,7 +599,7 @@ def run_node_pipeline(
         If squeeze_output=True and only one output then directly np.array.
     """
 
-    check_graph(nodes)
+    check_graph(nodes, check_for_peak_source=check_for_peak_source)
 
     job_kwargs = fix_job_kwargs(job_kwargs)
     assert all(isinstance(node, PipelineNode) for node in nodes)

src/spikeinterface/preprocessing/__init__.py

@@ -20,6 +20,8 @@
     PreprocessingPipeline,
 )
 
+from .detect_artifacts import detect_artifact_periods, detect_artifact_periods_by_envelope, detect_saturation_periods
+
 # for snippets
 from .align_snippets import AlignSnippets
 from warnings import warn