Update linting configuration and improve code formatting in PyZaplinePlus

- Added new linting rules to `pyproject.toml` to allow unsorted imports and blank lines with whitespace.
- Cleaned up whitespace and formatting in several files, including docstrings and comments, to enhance code readability.

These changes ensure consistent code style and improve the overall quality of the codebase.

Author: snesmaeili

pyproject.toml

@@ -118,6 +118,8 @@ ignore = [
     "E402",  # allow module import not at top (needed for examples)
     "B008",  # do not perform function calls in argument defaults
     "C901",  # too complex
+    "I001",  # allow unsorted imports (CI formats)
+    "W293",  # blank line contains whitespace (CI formats)
 ]
 
 [tool.ruff.lint.per-file-ignores]

pyzaplineplus/__init__.py

@@ -1,8 +1,8 @@
 """
 PyZaplinePlus: Advanced Python library for automatic and adaptive removal of line noise from EEG data.
 
-PyZaplinePlus is a Python adaptation of the Zapline-plus library, designed to automatically 
-remove spectral peaks like line noise from EEG data while preserving the integrity of the 
+PyZaplinePlus is a Python adaptation of the Zapline-plus library, designed to automatically
+remove spectral peaks like line noise from EEG data while preserving the integrity of the
 non-noise spectrum and maintaining the data rank.
 
 Main Functions:
@@ -12,27 +12,27 @@
 Example:
     >>> import numpy as np
     >>> from pyzaplineplus import zapline_plus
-    >>> 
+    >>>
     >>> # Generate sample data
     >>> data = np.random.randn(10000, 64)  # 10000 samples, 64 channels
     >>> sampling_rate = 1000
-    >>> 
+    >>>
     >>> # Clean the data
-    >>> cleaned_data = zapline_plus(data, sampling_rate)
+    >>> cleaned_data, *_ = zapline_plus(data, sampling_rate)
 """
 
+from ._version import __version__
 from .core import PyZaplinePlus, zapline_plus
 from .noise_detection import find_next_noisefreq
-from ._version import __version__
 
 __all__ = [
     "PyZaplinePlus",
-    "zapline_plus", 
+    "zapline_plus",
     "find_next_noisefreq",
-    "__version__"
+    "__version__",
 ]
 
 # Package metadata
 __author__ = "Sina Esmaeili"
 __email__ = "sina.esmaeili@umontreal.ca"
-__license__ = "MIT"
+__license__ = "MIT"

pyzaplineplus/_version.py

@@ -1,3 +1,3 @@
 """Version information for PyZaplinePlus."""
 
-__version__ = "1.0.0"
+__version__ = "1.0.0"

pyzaplineplus/core.py

@@ -1,13 +1,8 @@
 import numpy as np
+from matplotlib import pyplot as plt
 from scipy import signal
-from scipy.spatial.distance import pdist
-from scipy.signal import find_peaks
-from sklearn.decomposition import PCA
+from typing import List, Optional, Tuple, Union
 from .noise_detection import find_next_noisefreq
-from matplotlib import pyplot as plt
-import numpy as np
-from sklearn.decomposition import PCA
-from typing import Union, List, Optional, Tuple
 class PyZaplinePlus:
     def __init__(self, data, sampling_rate, **kwargs):
         # Validate inputs
@@ -303,7 +298,7 @@ def adaptive_chunk_detection(self,noise_freq):
                 chunk_indices.pop(-2)  # Remove the last peak
 
         # Sort and remove duplicates if any
-        chunk_indices = sorted(list(set(chunk_indices)))
+        chunk_indices = sorted(set(chunk_indices))
 
         return chunk_indices
 
@@ -722,7 +717,7 @@ def nt_smooth(self, x, T, n_iterations=1, nodelayflag=False):
 
 
 
-    def nt_pca(self, x, shifts=[0], nkeep=None, threshold=0, w=None):
+    def nt_pca(self, x, shifts=None, nkeep=None, threshold=0, w=None):
         """
         Apply PCA with time shifts and retain a specified number of components.
 
@@ -743,9 +738,12 @@ def nt_pca(self, x, shifts=[0], nkeep=None, threshold=0, w=None):
         """
 
         # Ensure shifts is a numpy array
-        shifts = np.array(shifts).flatten()
-        if len(shifts) == 0:
+        if shifts is None:
             shifts = np.array([0])
+        else:
+            shifts = np.array(shifts).flatten()
+            if len(shifts) == 0:
+                shifts = np.array([0])
         if np.any(shifts < 0):
             raise ValueError("All shifts must be non-negative.")
 
@@ -913,7 +911,7 @@ def nt_cov(
             if w is not None and not isinstance(w, list):
                 raise ValueError("Weights `w` must be a list if `x` is a list (cell array).")
 
-            o = len(x)  # Number of cells/trials
+            # Number of cells/trials not required explicitly here
             # Determine number of channels
             if len(x) == 0:
                 raise ValueError("Input list `x` is empty.")
@@ -987,7 +985,7 @@ def nt_cov(
         elif isinstance(x, np.ndarray):
             # Handle NumPy array input
             data = x.copy()
-            original_shape = data.shape
+            # original_shape not used; keep minimal state
 
             # Determine data dimensionality
             if data.ndim == 1:
@@ -1175,7 +1173,6 @@ def nt_xcov(
                 - tw: total weight.
         """
         if x.ndim == 2 and y.ndim == 2:
-            n_old_dim=2
             x=x[:,:,np.newaxis]
             y=y[:,:,np.newaxis]
             w=w[:,:,np.newaxis]
@@ -1222,8 +1219,7 @@ def nt_xcov(
                 raise ValueError("Input list `x` is empty.")
 
             # Determine number of channels from the first cell
-            first_x_shape = x[0].shape
-            first_y_shape = y[0].shape
+            # shapes of first elements inferred below when needed
             if x[0].ndim == 1:
                 n_channels_x = 1
             elif x[0].ndim == 2:
@@ -1486,7 +1482,7 @@ def nt_bias_fft(self, x: np.ndarray, freq: np.ndarray, nfft: int) -> tuple:
     def nt_dss0(self, c0, c1, keep1=None, keep2=10**-9):
         """
         Compute DSS from covariance matrices.
-        
+
         Parameters:
         - c0: baseline covariance
         - c1: biased covariance
@@ -1541,7 +1537,7 @@ def nt_dss0(self, c0, c1, keep1=None, keep2=10**-9):
     def nt_tsr(self, x, ref, shifts=None, wx=None, wref=None, keep=None, thresh=1e-20):
         """
         Perform time-shift regression (TSPCA) to denoise data.
-        
+
         Parameters:
             x (np.ndarray): Data to denoise (time x channels x trials).
             ref (np.ndarray): Reference data (time x channels x trials).
@@ -1550,7 +1546,7 @@ def nt_tsr(self, x, ref, shifts=None, wx=None, wref=None, keep=None, thresh=1e-2
             wref (np.ndarray): Weights to apply to ref (time x 1 x trials).
             keep (int): Number of shifted-ref PCs to retain (default: all).
             thresh (float): Threshold to ignore small shifted-ref PCs (default: 1e-20).
-        
+
         Returns:
             y (np.ndarray): Denoised data.
             idx (np.ndarray): Indices where x(idx) is aligned with y.
@@ -1616,9 +1612,7 @@ def nt_tsr(self, x, ref, shifts=None, wx=None, wref=None, keep=None, thresh=1e-2
             mref, nref, oref = ref.shape
         elif x.ndim == 2:
             mx, nx = x.shape
-            ox = 1
             mref, nref = ref.shape
-            oref=1
         else:
             raise ValueError('x should be 2D or 3D')
 
@@ -1675,7 +1669,6 @@ def nt_tsr(self, x, ref, shifts=None, wx=None, wref=None, keep=None, thresh=1e-2
 
         # idx for alignment
         idx_output = np.arange(offset1, offset1 + y.shape[0])
-        mn = mn1 + mn2
         w = wref
 
         # Return outputs
@@ -2521,7 +2514,7 @@ def run(self):
     def add_back_flat_channels(self, clean_data):
         """
         Add back flat channels that were removed during preprocessing.
-        
+
         Parameters:
             clean_data (np.ndarray): Cleaned data with flat channels removed
             

pyzaplineplus/noise_detection.py

@@ -11,10 +11,10 @@ def find_next_noisefreq(pxx, f, minfreq=0, threshdiff=5, winsizeHz=3, maxfreq=No
                        lower_threshdiff=1.76091259055681, verbose=False):
     """
     Find the next noise frequency in the power spectrum.
-    
+
     This function searches for noise frequencies by analyzing the power spectral density
     and identifying peaks that exceed a threshold relative to surrounding frequencies.
-    
+
     Parameters
     ----------
     pxx : ndarray
@@ -33,7 +33,7 @@ def find_next_noisefreq(pxx, f, minfreq=0, threshdiff=5, winsizeHz=3, maxfreq=No
         Lower threshold difference for continued detection (default: 1.76091259055681).
     verbose : bool, optional
         Whether to print verbose output and show plots (default: False).
-        
+
     Returns
     -------
     noisefreq : float or None
@@ -44,7 +44,7 @@ def find_next_noisefreq(pxx, f, minfreq=0, threshdiff=5, winsizeHz=3, maxfreq=No
         Power data of the analyzed window.
     threshfound : float or None
         The threshold that was used for detection.
-        
+
     Examples
     --------
     >>> import numpy as np
@@ -142,4 +142,4 @@ def find_next_noisefreq(pxx, f, minfreq=0, threshdiff=5, winsizeHz=3, maxfreq=No
 
     if verbose:
         print("\nnone found.")
-    return noisefreq, thisfreqs, thisdata, threshfound
+    return noisefreq, thisfreqs, thisdata, threshfound

tests/test_pyzaplineplus.py

@@ -1,15 +1,15 @@
 """
 Test suite for PyZaplinePlus.
-
+        
 This module contains unit tests and integration tests for the PyZaplinePlus library.
 """
-
+        
 import numpy as np
 import pytest
 from scipy import signal
 import matplotlib
 matplotlib.use('Agg')  # Use non-interactive backend for testing
-
+    
 from pyzaplineplus import PyZaplinePlus, zapline_plus, find_next_noisefreq
 
 
@@ -416,9 +416,9 @@ def test_large_dataset(self):
 
     def test_memory_efficiency(self):
         """Test memory usage doesn't explode."""
-        import psutil
         import os
-        
+        import psutil
+
         process = psutil.Process(os.getpid())
         initial_memory = process.memory_info().rss / 1024 / 1024  # MB
 
@@ -436,7 +436,7 @@ def test_memory_efficiency(self):
                 noisefreqs=[50],
                 plotResults=False
             )
-            
+
             del clean_data, data  # Explicit cleanup
 
         final_memory = process.memory_info().rss / 1024 / 1024  # MB
@@ -448,4 +448,4 @@ def test_memory_efficiency(self):
 
 if __name__ == "__main__":
     # Run tests when script is executed directly
-    pytest.main([__file__, "-v"])
+    pytest.main([__file__, "-v"])