Other data structures, functions, and types

class radiens.lib.channel_metadata.ChannelMetadata(raw=None)

Container for channel metadata.

property table: Metadata table providing metadata on each channel (aka signal) in the dataset.

property attributes: Dictionary of attributes

index(stype)

Signal indices by signal type and index type. All indices are zero-indexed.

Parameters:: stype (SignalType) – requested signal type

Examples

>>> meta.index[SignalType.AMP]
>>> KEY_IDXS(ntv=[5,1,23,46], dset=[0,1,2,3], sys=[5,1,23,46])
>>> meta.index[SignalType.AIN]
>>> KEY_IDXS(ntv=[3,2,0,1], dset=[0,1,2,3], sys=[35,34,32,33])  # signal has 32 AMP channels

Notes:

Index definitions:

ntv is the position of the signal (channel) trace when the backing dataset was created for each signal type. For recording files, it is the position of the channels in the DAQ data stream or primary data file. The ntv index starts at zero for each signal type.

dset is the position of a signal trace in the backing dataset for each signal type. Unless the signal traces in the dataset were re-ordered, dset will equal ntv. The dset index starts at zero for each signal type.

sys is the position of the signal (channel) trace when the backing dataset was created. The sys index starts at zero and increases in natural order over all signal types alway in the following order: ‘amp’, ‘gpio_ain’, ‘gpio_din’, ‘gpio_dout’.

sel_index(stype)

Selected signal indices by signal type and index type.

See index() for more information on signal indices.

num_sigs(stype): Number of signals for the requested signal type.

sel_num_sigs(stype): Number of selected signals for the requested signal type.

num_sigs_total(): Total number of signals

property num_selected_sigs_total: Total number of signals

property sensor_spec: Dict with ports (str) as keys mapping to dicts describing headstage, site, and probe wireframe metadata

class radiens.lib.dataset_metadata.DatasetMetadata(raw_msg=None)

property id: Dataset ID of this DataSource the Radiens hub. None indicates that it is not on a hub.

property time_range: Dataset time range

property TR: Aliases obj.time_range

property channel_metadata: Dataset’s ChannelMetadata, which describes the dataset’s signals (aka channels), probes, positions, etc.

property attributes: Dataset attributes.

property path: File system path of the Radiens data file set that backs this DataSource.

property base_name: File system base name of the Radiens data file set that backs this DataSource.

property table: Dataset attributes as a table.

clear_dataset_id(): Clears this DataSource’s dataset ID by setting it to None. This is a power user function. Best practice is to call it when the dataset has been cleared from the Radiens hub.

class radiens.lib.allego_lib.AllegoState(raw=<class 'allegoserver_pb2.ConfigAndStatus'>)

Container for Allego configuration

property sample_freq: Current system sample frequency in samples/sec [Hz]

property system_mode: Current Allego mode, e.g. ‘sim-spikes’, etc.

property port_num_channels: Number of channels on each connected port.

property cable_length_ft: Headstage cable length in feet for each system port

property recording: Recording status

property stream: Streaming status

property connected: Returns true if there is a headstage connected on any channel.

property base_name: Returns true if there is a headstage connected on any channel.

class radiens.lib.signals_snapshot.Signals(raw)

Container for a multi-channel (multi-trace) dataset for amplifier, GPIO analog in, GPIO digital in, and GPIO digital out signals.

All traces for all signal types have the same sampling frequency and start and end times. Thus, the set of traces of each signal type is a 2-D matrix.

property time_range: dataset time range in seconds

property TR: Aliases dataset time range.

property signals

Time-series signals as a named tuple by signal type.

For each signal type, the multi-channel dataset is a 2-D numpy array, with dim 0=trace position and dim 1 is the sample values over the time range.

property attributes: Dataset attributes

property channel_metadata: Dataset channel metadata

class radiens.lib.sig_metrics.SignalMetrics(dsource_id, resp=<class 'common_pb2.KpiBundlePacketMetrics'>)

Container for signal metrics

property time_range: Signal metrics time range

property TR: Aliases signal metrics time range

property ntv_idxs: Signal metrics native channel indices

property settings: Signal metrics settings (dict)

property metric_ids: Ordered list of signal metric IDs.

property packet_idxs: Signal metrics packet indices

property num_packets: Signal metrics number of packets

property num_sigs: Signal metrics number of signals

property packet_dur_sec: Signal metrics packet duration in seconds

property val

Signal metrics settings as multi-dimensional np.ndarray, with dim 0=packet idx, dim 1=channel index, dim 2=metric index. The ordered metric IDs are available in obj.metric_ids.

Examples:

access packet 20, channel 23, and metric 1
val[20][23][1]
access channel 23, metric 1 over all packets:
val[:,23,1]
access metric 1 over all packets and channels:
z = y[:,:,1] # returns 2-D array, where dim 0=packet index and dim 1=channels np.ravel(z) # returns 1-D array with packets flattened

property packet_stats

print(): Prints a summary of the metrics.

class radiens.lib.sig_metrics.METRIC_ID(mode, name)

Example

>>> metric_id = METRIC_ID(mode=METRIC_MODE.BASE, name=METRIC.MEAN)

mode: Alias for field number 0

name: Alias for field number 1

class radiens.lib.sig_metrics.METRIC_MODE(value)

An enumeration.

BASE = 0: base packet

AVG = 1: running average of all base packets in the stream

STREAM = 2: cumulative over all base packets in the stream

class radiens.lib.sig_metrics.METRIC(value)

An enumeration.

NUM_PTS = 0: Number of points in packet

NUM_PTS_ISI = 1: Number of interspike intervals in packet

DUR_SEC = 2: Packet duration in seconds

MEAN = 3: Mean signal level in packet

MIN = 4: Minimum signal level in packet

MIN_ISI = 5: Minimum interspike signal level in packet

MAX = 6: Maximum signal level in packet

MAX_ISI = 7: Maximum interspike signal level in packet

MAX_ABS = 8: Absolute maximum signal level in packet

MAX_ABS_ISI = 9: Absolute maximum interspike signal level in packet

TIMESTAMP_MIN = 10: Timestamp of minimum signal level in packet

TIMESTAMP_MAX = 11: Timestamp of maximum signal level in packet

MAX_MIN_DIFF_ABS = 12: Absolute difference between maximum and minimum signal levels in packet

MAX_MIN_DIFF_ABS_ISI = 13: Absolute difference between maximum and minimum interspike signal levels in packet

SD = 14: Signal standard deviation in packet

SD_ISI = 15: Signal standard deviation calculated over interspike intervals

VAR = 16: Signal variance in packet

VAR_ISI = 17: Signal variance calculated over interspike intervals

RMS = 18: Root mean square (RMS) signal value in packet

RMS_ISI = 19: RMS signal value calculated over interspike intervals

NOISE_UV = 20: Noise level in microvolts

SNR = 21: Signal-to-noise ratio (SNR) of events in packet

NUM_EVENTS = 22: Number of events in packet

EVENT_RATE = 23: Event rate in packet

EVENT_MAX = 24: Maximum amplitude of events in packet

EVENT_MIN = 25: Minimum amplitude of events in packet

EVENT_MAX_ABS = 26: Absolute maximum amplitude of events in packet

EVENT_MAX_MIN_DIFF_ABS = 27: Absolute difference between maximum and minimum amplitudes of events in packet

EVENT_TIMESTAMP_MIN = 28: Timestamp of minimum amplitude of events in packet

EVENT_TIMESTAMP_MAX = 29: Timestamp of maximum amplitude of events in packet

EVENT_TIMESTAMP_MAX_ABS = 30: Timestamp of maximum absolute amplitude of events in packet

EVENT_TIMESTAMP_MAX_MIN_DIFF_ABS = 31: Timestamp of absolute difference between maximum and minimum amplitudes of events in packet

MEAN_MAX = 32: Mean peak maximum amplitude of events in packet (max over all base packets)

MEAN_MIN = 33: Mean peak minimum amplitude of events in packet (min over all base packets)

MEAN_MAX_ABS = 34: Mean peak absolute maximum amplitude of events in packet (max over all base packets)

EVENT_MEAN_MAX_MIN_DIFF_ABS = 35: Mean peak absolute amplitude of events in packet (max over all base packets)

radiens.utils.util.make_time_range(time_range=None, timestamp=None, fs=None, walltime=None, pbTR=None)

Utility function to compose the TIME_RANGE named tuple composed from either a time range or timestamp list.

Parameters:

time_range (list | numpy.ndarray) – time range in seconds as [time start, time end] (optional, default=None).
timestamp (list | numpy.ndarray) – timestamp in seconds as [timestamp start, timestamp end] (optional, default=None).
fs (float) – sampling frequency in samples/sec (required, default=None)
walltime (datetime.datetime) – wall time of the time range start.
pbTR – low-level system parameter not available to user.

Notes:

TIME_RANGE named tuple fields:: sec (numpy.ndarray): [time start, time end] timestamp (numpy.ndarray): [timestamp start, timestamp end] fs (float): sampling frequency walltime: wall time of the time range start dur_sec (float): imputed time range duration equal to TR.time_range[1]-TR.time_range[0]. N (int): imputednumber of sample points equal to TR.timestamp[1]-TR.timestamp[0].

Either timestamp or time_range must be provided, with timestamp having preference. walltime is optional. dur_sec and N are imputed from the arguments.

System use only (not available to user) Either pbTR or (fs and either time_range or timestamp) must be provided as arguments. pbTR has preference.

class radiens.utils.constants.KeyIndex(value)

An enumeration of key index types

Example

>>> key_ind = KeyIndex.NTV

Valid values are as follows:

NTV = 0: native

DATA = 1: data

SYS = 2: system

class radiens.utils.constants.SignalType(value)

An enumeration of signal types

Example

>>> sig_type = SignalType.AMP

Valid values are as follows:

AMP = 0: primary signals

AIN = 1: analog in

DIN = 2: digital in

DOUT = 3: digital out

class radiens.utils.constants.TRS_MODE(value)

TRS_MODE is the time range selector (TRS) enum used to control time range selection in calls to get signals, spikes, power spectral density, or similar data sets.

Example

>>> sel_mode = TRS_MODE.SUBSET

Given a time range [start, end) in seconds (dataset time), the selector modes are:

FROM_HEAD = 2: selects [(head of stream/cache/file - start)

SUBSET = 0: selects [start, end)

TO_HEAD = 1: selects [start, head of stream/cache/file]

class radiens.utils.constants.PSD_SCALING(value): PSD_SCALING is used to set the power spectral density (PSD) scale.

class radiens.utils.constants.FFT_WINDOW(value): FFT_WINDOW is used to set the time-domain window function for power spectral density (PSD) analysis.

class radiens.utils.constants.TIME_RANGE(sec, timestamp, fs, dur_sec, walltime, N)

N: Alias for field number 5

dur_sec: Alias for field number 3

fs: Alias for field number 2

sec: Alias for field number 0

timestamp: Alias for field number 1

walltime: Alias for field number 4