ioiopype 0.0.7

A realtime processing framework for python

IOIOpype

IOIOpype is processing framework for realtime applications written in python. Data is propergated between Nodes that can be connected via Streams. Nodes can be input nodes 'INode', output nodes 'ONode' or input and output nodes 'IONode'. Algorithms and signal processing pipelines can be prototyped easily by combining multiple nodes.

Supported Devices

• Unicorn Hybrid Black (g.tec medical engineering GmbH)
• Unicorn Hybrid Black Simulator

INodes

• ConsoleLog - Writes received data to the console
• ToWorkspace - Propagates data to the main thread via event
• FramePlot - Plots a data frame
• SamplePlot - Plots time series data
• CSV Logger - Logs data to csv files

IONodes

• Buffer - Buffers data with a defined window size and overlap
• PWelch - Calculates a Pwelch spectrum from a frame
• Downsample - Samples a data frame down by a given factor
• OffsetCorrection - Removes the offset from a data frame
• ButterworthFilter - Applies a butterworth filter to a sample based signal
• ToSample - Slices and forwards a data frame into samples
• Log - Applies a logarithm to the input signal (ln, log10, 10*log10)
• Square - Squares the input signal
• Sqrt - Calsulates the square root of the input signal
• Mux - Concatenates multiple signals into one
• Operation - Adds, Subtracts, Multiplies, Divides or applies a matrix multiplication

ONodes

• Constant - forwards constant data with a defined sampling rate
• NoiseGenerator - Generates gaussian noise as a timeseries signal.
• SignalGenerator - Generates sample based timeseries signals. Sine, Square, Sawtooth and Triangle signals can be generated
• Frame - Allows to send a data frame to INodes

Supported platforms

Windows	Linux	Mac

Install ioiopype from PyPi

pip install ioiopype

Unicorn Hybrid Black - Acquisition and data visualization example

This example shows how to connect to a 'Unicorn - The Brain Interface' device and establish a data acquisition. Raw data acquired from the device is filtered with Butterworth IIR filters and visualized using a timeseries plot. Additionally the raw spectrum is calculated using a pwelch spectrum. The spectrum is visualized using a frame plot.

from PySide6.QtWidgets import QApplication
import ioiopype as ioio

app = QApplication(sys.argv)

use_device_simulator = True #Use device simulator (True) or real device (False)
if use_device_simulator:
    device = ioio.UnicornSimulator('UN-0000.00.00')
else:
    device = ioio.Unicorn('UN-2023.02.15') #Enter your device serial here

#initialize processing nodes
buf = ioio.Buffer(device.NumberOfEEGChannels, 4 * device.SamplingRateInHz, 4 * device.SamplingRateInHz - 25)
pw = ioio.PWelch(device.SamplingRateInHz)
fp = ioio.FramePlot(samplingRate=4)
hp = ioio.ButterworthFilter(ioio.FilterType.Highpass, device.SamplingRateInHz, 2, [2])
lp = ioio.ButterworthFilter(ioio.FilterType.Lowpass, device.SamplingRateInHz, 4, [30])
n50 = ioio.ButterworthFilter(ioio.FilterType.Notch, device.SamplingRateInHz, 4, [48, 52])
n60 = ioio.ButterworthFilter(ioio.FilterType.Notch, device.SamplingRateInHz, 4, [58, 62])
sp = ioio.SamplePlot(device.NumberOfEEGChannels, device.SamplingRateInHz, 6, 100)

#build ioiopype
device.connect(0, buf.InputStreams[0])
buf.connect(0, pw.InputStreams[0])
pw.connect(0, fp.InputStreams[0])

device.connect(0, n50.InputStreams[0])
n50.connect(0, n60.InputStreams[0])
n60.connect(0, hp.InputStreams[0])
hp.connect(0, lp.InputStreams[0])
lp.connect(0, sp.InputStreams[0])

app.exec()

#disconnect ioiopype
device.disconnect(0, buf.InputStreams[0])
buf.disconnect(0, pw.InputStreams[0])
pw.disconnect(0, fp.InputStreams[0])

device.disconnect(0, n50.InputStreams[0])
n50.disconnect(0, n60.InputStreams[0])
n60.disconnect(0, hp.InputStreams[0])
hp.disconnect(0, lp.InputStreams[0])
lp.disconnect(0, sp.InputStreams[0])

#close device
del device

Contact

Support this project:
Join IOIOPype on Discord
Contact: mwalchsoferyt at gmail dot com