cleaned up unused imports

used ruff linter to clean code

Tests/test_PseudoDevice.py

@@ -1,7 +1,5 @@
 import time
-import pytest
 from pybci import PyBCI
-from pybci.Utils.PseudoDevice import PseudoDeviceController
 # Test case using the fixture
 #@pytest.mark.timeout(300)  # Extended timeout to 5 minutes
 def test_run_pseudo():

Tests/test_Pytorch.py

@@ -1,7 +1,5 @@
 from pybci import PyBCI
 import time
-import pytest
-
 import torch
 from torch.utils.data import DataLoader, TensorDataset
 from torch import nn

Tests/test_Simple.py

@@ -1,7 +1,5 @@
 from pybci import PyBCI
-import time, os
-import pytest
-
+import time
 # Test case using the fixture
 #@pytest.mark.timeout(300)  # Extended timeout to 5 minutes
 def test_run_bci():

Tests/test_Sklearn.py

@@ -1,6 +1,5 @@
 from pybci import PyBCI
 import time
-import pytest
 from sklearn.neural_network import MLPClassifier
 # Test case using the fixture
 #@pytest.mark.timeout(300)  # Extended timeout to 5 minutes

Tests/test_Tensorflow.py

@@ -1,7 +1,5 @@
 from pybci import PyBCI
 import time
-import pytest
-
 import tensorflow as tf# bring in tf for custom model creation
 
 num_chs = 8 # 8 channels are created in the PseudoLSLGenerator

appveyor.yml

@@ -1,4 +1,4 @@
-version: 1.0.{build}
+version: 1.2.{build}
 image: Visual Studio 2019
 environment:
   matrix:

pybci/CliTests/testPyTorch.py

@@ -1,14 +1,13 @@
-import time, argparse
+import time
+import argparse
 from pybci import PyBCI
-from pybci.Configuration.EpochSettings import GlobalEpochSettings
-
 import torch
 from torch.utils.data import DataLoader, TensorDataset
 from torch import nn
 
 
 def main(create_pseudo_device=True, min_epochs_train=4, min_epochs_test=10, num_chs = 8, num_feats = 2, num_classes = 4):
-    if create_pseudo_device == True:
+    if create_pseudo_device:
         num_chs = 8 # 8 channels are created in the PseudoLSLGenerator
         num_feats = 2 # default is mean freq and rms to keep it simple
         num_classes = 4 # number of different triggers (can include baseline) sent, defines if we use softmax of binary

pybci/CliTests/testSklearn.py

@@ -1,6 +1,6 @@
 from pybci import PyBCI
-from pybci.Configuration.EpochSettings import GlobalEpochSettings
-import time, argparse
+import time
+import argparse
 from sklearn.neural_network import MLPClassifier
 
 def main(create_pseudo_device=True, min_epochs_train=4, min_epochs_test=10):

pybci/CliTests/testTensorflow.py

@@ -1,11 +1,10 @@
-import time, argparse
+import time
+import argparse
 from pybci import PyBCI
-from pybci.Configuration.EpochSettings import GlobalEpochSettings
 import tensorflow as tf# bring in tf for custom model creation
 
-
 def main(create_pseudo_device=True, min_epochs_train=4, min_epochs_test=10, num_chs = 8, num_feats = 2, num_classes = 4):
-    if create_pseudo_device == True:
+    if create_pseudo_device:
         num_chs = 8 # 8 channels are created in the PseudoLSLGenerator
         num_feats = 2 # default is mean freq and rms to keep it simple
         num_classes = 4 # number of different triggers (can include baseline) sent, defines if we use softmax of binary

pybci/Examples/ArduinoHandGrasp/testArduinoHand.py

@@ -1,7 +1,6 @@
 import time
 from pybci import PyBCI
 from pybci.Configuration.EpochSettings import GlobalEpochSettings
-from pybci.Utils.FeatureExtractor import GenericFeatureExtractor, GeneralFeatureChoices
 from pybci.Utils.Logger import Logger
 import numpy as np
 from scipy.fft import fft

pybci/Examples/ArduinoHandGrasp/testArduinoPytorch.py

@@ -1,12 +1,11 @@
-import time, torch
+import time
+import torch
 from torch.utils.data import DataLoader, TensorDataset
 from torch import nn
 from pybci import PyBCI
 from pybci.Configuration.EpochSettings import GlobalEpochSettings
-from pybci.Utils.FeatureExtractor import GenericFeatureExtractor, GeneralFeatureChoices
 from pybci.Utils.Logger import Logger
 import numpy as np
-from scipy.fft import fft
 # We control the arduino via lsl python script which is responsible for COM connection
 from pylsl import StreamInfo, StreamOutlet
 # Set up the LSL stream info

pybci/Examples/MultimodalPupilLabsEEG/testMultimodal.py

@@ -1,7 +1,7 @@
 import time
 from pybci import PyBCI
 import numpy as np
-from pybci.Utils.FeatureExtractor import GenericFeatureExtractor
+from scipy.fft import fft
 
 class PupilGazeDecode():
     # pupil-labs channels:
@@ -24,7 +24,7 @@ def ProcessFeatures(self, epochData, sr, epochNum): # This is the required funct
             bothmean = np.mean([(epochData[1][i] + epochData[2][i]) / 2 for i in range(len(epochData[1]))]) # mean of both eyes in 3d
             return np.nan_to_num([confidence, rightmean,leftmean,bothmean]) #  expects 2d
 
-from scipy.fft import fft
+
 class EOGClassifier():
     # used Fp1 and Fp2 from io:bio EEG device
     def ProcessFeatures(self, epochData, sr, epochNum): # Every custom class requires a function with this name and structure to extract the featur data and epochData is always [Samples, Channels]

pybci/Examples/testEpochTimingsConfig.py

@@ -1,6 +1,6 @@
 import time
 from pybci import PyBCI
-from pybci.Configuration.EpochSettings import GlobalEpochSettings, IndividualEpochSetting
+from pybci.Configuration.EpochSettings import GlobalEpochSettings
 
 gs = GlobalEpochSettings()
 gs.tmax = 1 # grab 1 second after marker

pybci/Examples/testPyTorch.py

@@ -1,7 +1,5 @@
 import time
 from pybci import PyBCI
-from pybci.Configuration.EpochSettings import GlobalEpochSettings
-
 import torch
 from torch.utils.data import DataLoader, TensorDataset
 from torch import nn

pybci/Examples/testRaw.py

@@ -1,9 +1,10 @@
-import torch, time
+import torch
+import time
 from torch.utils.data import DataLoader, TensorDataset
 from torch import nn
 from pybci import PyBCI
 import numpy as np
-from pybci.Utils.Logger import Logger
+
 num_chs = 3 # 8 channels are created in the PseudoLSLGenerator, but we drop 5 to save compute (real-time CNN can be computationally heavy!)
 sum_samps = 125 # sample rate is 250 in the PseudoLSLGwnerator
 num_classes = 4 # number of different triggers (can include baseline) sent, defines if we use softmax of binary

pybci/Examples/testSklearn.py

@@ -1,5 +1,4 @@
 from pybci import PyBCI
-from pybci.Configuration.EpochSettings import GlobalEpochSettings
 import time
 from sklearn.neural_network import MLPClassifier
 clf = MLPClassifier(max_iter = 1000, solver ="lbfgs")#solver=clf, alpha=alpha,hidden_layer_sizes=hid)

pybci/Examples/testTensorflow.py

@@ -1,6 +1,5 @@
 import time
 from pybci import PyBCI
-from pybci.Configuration.EpochSettings import GlobalEpochSettings
 import tensorflow as tf# bring in tf for custom model creation
 
 num_chs = 8 # 8 channels are created in the PseudoLSLGenerator

pybci/ThreadClasses/AsyncDataReceiverThread.py

@@ -1,6 +1,5 @@
-import threading, time
+import threading
 from collections import deque
-import itertools
 from bisect import bisect_left
 def slice_fifo_by_time(fifo, start_time, end_time):
         """Find the slice of fifo between start_time and end_time using binary search."""
@@ -48,7 +47,7 @@ def run(self):
         dataFIFOs = [deque(maxlen=fifoLength) for ch in range(chCount - len(self.streamChsDropDict))]
         while not self.closeEvent.is_set():
             sample, timestamp = self.dataStreamInlet.pull_sample(timeout = 1)
-            if sample != None:
+            if sample is not None:
                 for index in sorted(self.streamChsDropDict, reverse=True):
                     del sample[index] # remove the desired channels from the sample
                 for i,fifo in enumerate(dataFIFOs):
@@ -114,7 +113,7 @@ def run(self):
                 # add levels of debug?
 
     def ReceiveMarker(self, marker, timestamp): # timestamp will be used for non sample rate specific devices (pupil-labs gazedata)
-        if self.startCounting == False: # only one marker at a time allow, other in windowed timeframe ignored
+        if self.startCounting is False: # only one marker at a time allow, other in windowed timeframe ignored
             self.currentMarker = marker
             self.markerTimestamp = timestamp
             if len(self.customEpochSettings.keys())>0: #  custom marker received

pybci/ThreadClasses/ClassifierThread.py

@@ -1,6 +1,8 @@
 from ..Utils.Classifier import Classifier 
 from ..Utils.Logger import Logger
-import queue,threading, time
+import queue
+import threading
+import time
 import numpy as np
 
 class ClassifierThread(threading.Thread):

pybci/ThreadClasses/DataReceiverThread.py

@@ -33,7 +33,7 @@ def run(self):
         dataFIFOs = [deque(maxlen=fifoLength) for ch in range(chCount - len(self.streamChsDropDict))]
         while not self.closeEvent.is_set():
             sample, timestamp = self.dataStreamInlet.pull_sample(timeout = 1)
-            if sample != None:
+            if sample is not None:
                 for index in sorted(self.streamChsDropDict, reverse=True):
                     del sample[index] # remove the desired channels from the sample
                 for i,fifo in enumerate(dataFIFOs):
@@ -88,7 +88,7 @@ def run(self):
                 # add levels of debug 
 
     def ReceiveMarker(self, marker, timestamp): # timestamp will be used for non sample rate specific devices (pupil-labs gazedata)
-        if self.startCounting == False: # only one marker at a time allow, other in windowed timeframe ignored
+        if self.startCounting is False: # only one marker at a time allow, other in windowed timeframe ignored
             self.currentMarker = marker
             if len(self.customEpochSettings.keys())>0: #  custom marker received
                 if marker in self.customEpochSettings.keys():

pybci/ThreadClasses/FeatureProcessorThread.py

@@ -1,8 +1,9 @@
-import threading, queue, time
+import threading
+import queue
+import time
 from ..Utils.FeatureExtractor import GenericFeatureExtractor
 from ..Utils.Logger import Logger
 from ..Configuration.EpochSettings import GlobalEpochSettings
-import copy
 
 class FeatureProcessorThread(threading.Thread):
     tempDeviceEpochLogger = []
@@ -43,9 +44,9 @@ def run(self):
                     features = self.featureExtractor.ProcessFeatures(dataFIFOs, sr, target) # allows custom epoch class to be passed
                     if (self.logger.level == Logger.TIMING):
                         end = time.time()
-                        self.logger.log(Logger.TIMING, f" Feature Extraction time {end - start}")
+                        self.logger.log(Logger.TIMING, " Feature Extraction time "+str(end - start))
                         if (end-start) >self.globalWindowSettings.windowLength:
-                            self.logger.log(Logger.WARNING, f" Feature Extraction time > globalEpochSetting.windowLength, will create lag in classification output. Recommended to reduce channels, smapling rate, and features or reduce feature computational complexity.")
+                            self.logger.log(Logger.WARNING, " Feature Extraction time > globalEpochSetting.windowLength, will create lag in classification output. Recommended to reduce channels, smapling rate, and features or reduce feature computational complexity.")
                     self.featureQueueTrain.put( [features, devCount, target, dict(self.epochCounts)] )
                 except queue.Empty:
                     pass
@@ -56,9 +57,9 @@ def run(self):
                     features = self.featureExtractor.ProcessFeatures(dataFIFOs, sr, None)
                     if (self.logger.level == Logger.TIMING):
                         end = time.time()
-                        self.logger.log(Logger.TIMING, f" Feature Extraction time {end - start}")
+                        self.logger.log(Logger.TIMING, " Feature Extraction time "+str(end - start))
                         if (end-start) >self.globalWindowSettings.windowLength:
-                            self.logger.log(Logger.WARNING, f" Feature Extraction time > globalEpochSetting.windowLength, will create lag in classification output. Recommended to reduce channels, smapling rate, and features or reduce feature computational complexity.")
+                            self.logger.log(Logger.WARNING, " Feature Extraction time > globalEpochSetting.windowLength, will create lag in classification output. Recommended to reduce channels, smapling rate, and features or reduce feature computational complexity.")
                     self.featureQueueTest.put([features, devCount])
                 except queue.Empty:
                     pass

pybci/ThreadClasses/MarkerThread.py

@@ -16,7 +16,7 @@ def run(self):
         while not self.closeEvent.is_set():
             marker, timestamp = self.markerStreamInlet.pull_sample(timeout = 10)
             if self.trainTestEvent.is_set(): # We're training!
-                if marker != None:
+                if marker is not None:
                     marker = marker[0]
                     for thread in self.dataThreads:
                         thread.ReceiveMarker(marker, timestamp)

pybci/ThreadClasses/OptimisedDataReceiverThread.py

@@ -1,8 +1,5 @@
-import threading, time
-from collections import deque
-import itertools
+import threading
 import numpy as np
-from bisect import bisect_left
 
 class OptimisedDataReceiverThread(threading.Thread):
     """Responsible for receiving data from accepted LSL outlet, slices samples based on tmin+tmax basis, 
@@ -125,7 +122,7 @@ def run(self):
                 # add levels of debug?
 
     def ReceiveMarker(self, marker, timestamp): # timestamp will be used for non sample rate specific devices (pupil-labs gazedata)
-        if self.markerReceived == False: # only one marker at a time allow, other in windowed timeframe ignored
+        if self.markerReceived is False: # only one marker at a time allow, other in windowed timeframe ignored
             self.currentMarker = marker
             self.markerTimestamp = timestamp
             if len(self.customEpochSettings.keys())>0: #  custom marker received

pybci/Utils/Classifier.py

@@ -1,7 +1,6 @@
 import sklearn
 from sklearn.preprocessing import StandardScaler
 from sklearn import svm
-import tensorflow
 import torch
 
 from sklearn.model_selection import train_test_split
@@ -16,20 +15,20 @@ class Classifier():
 
     def __init__(self, clf = None, model = None, torchModel = None):
         super().__init__()
-        if clf != None:
+        if clf is not None:
             self.clf = clf
-        elif model != None:
+        elif model is not None:
             self.model = model
-        elif torchModel != None:
+        elif torchModel is not None:
             self.torchModel = torchModel
         self.CheckClassifierLibrary()
 
     def CheckClassifierLibrary(self):
-        if self.model != None: # maybe requires actual check for tensorflow model
+        if self.model is not None: # maybe requires actual check for tensorflow model
             self.classifierLibrary = "tensor"
-        elif self.torchModel != None: # maybe requires actual check for sklearn clf
+        elif self.torchModel is not None: # maybe requires actual check for sklearn clf
             self.classifierLibrary = "pyTorch"
-        elif self.clf != None: # maybe requires actual check for sklearn clf
+        elif self.clf is not None: # maybe requires actual check for sklearn clf
             self.classifierLibrary = "sklearn"
 
     def TrainModel(self, features, targets):