chapter1_04_time_series.md

type
slides

Adding data

---

Adding time series data

Neurophysiology data often consists of values that vary over time, or time series data, such as voltages, fluorescence intensities, or behavioral measures. PyNWB stores time series data using the TimeSeries class and its specialized subclasses. To create a TimeSeries object, you need to pass the following as arguments to the constructor: a name, the data, the unit of measurement, and timestamps for each data point, in seconds.

Let's create a TimeSeries object to represent a mouse's running speed, in meters/second, using randomly generated values for 1000 time points.

from pynwb import TimeSeries
from numpy.random import rand # to create some random data

# Generate some random data
data = rand(1000)
timestamps = list(range(1000))

# Create a TimeSeries object
running_speed = TimeSeries(name='RunningSpeed',
                           data=data,
                           unit='m/s',
                           timestamps=timestamps)

---

Adding the data to the NWBFile

Use the NWBFile method add_acquisition to add your time series to the NWBFile and label it as acquisition data.

nwbfile.add_acquisition(running_speed)

---

Accessing acquisition data of the NWBFile

Accessing data stored in the NWBFile is about as easy as accessing metadata stored in the NWBFile. Since an NWBFile can have multiple acquisition data, use the NWBFile method get_acquisition and pass the name of the TimeSeries that you want to access.

running_speed_read = nwbfile.acquisition['RunningSpeed']
print('First running speed:', running_speed_read.data[0], running_speed_read.unit)

First running speed: 0.5596276463909797 m/s

Inspecting a NWBFile

We can inspect the NWBFile object at a high level by simply printing out the object.

print(nwbfile)

root <class 'pynwb.file.NWBFile'>
Fields:
  acquisition: { RunningSpeed <class 'pynwb.base.TimeSeries'> }
  analysis: { }
  devices: { }
  electrode_groups: { }
  epoch_tags: {}
  ic_electrodes: { }
  imaging_planes: { }
  lab_meta_data: { }
  ogen_sites: { }
  processing_modules: { }
  stimulus: { }
  stimulus_template: { }
  time_intervals: { }

---

Let's practice!