radiens_core.curate_client

CurateClient — data curation client.

Attributes

Classes

CurateClient

Bases: BaseFileClient

Client for data curation operations.

Auto-discovers a running radiensserver on the local machine.

Usage::

with CurateClient() as client:
    meta = client.link_data_file("/path/to/recording.xdat")
    new_meta = client.slice_time(meta, 0, 10, "sliced.xdat")

Functions

apply_protocol

apply_protocol(protocol_id)

Re-apply a previously registered protocol by ID.

Parameters:

Name	Type	Description	Default
`protocol_id`	`str`	The ID of a protocol already registered with the server (e.g., returned from a previous transform call).	required

bandpass

bandpass(
    dataset, low_freq, high_freq, output_path, *, order=None
)

Apply a band-pass filter to a dataset.

Parameters:

Name	Type	Description	Default
`dataset`	`FlexDataset`	Dataset to filter (metadata or ID).	required
`low_freq`	`float`	Lower cutoff frequency in Hz.	required
`high_freq`	`float`	Upper cutoff frequency in Hz.	required
`output_path`	`str \| Path`	Where to save the result.	required
`order`	`int \| None`	Filter order. If `None`, the server default is used.	`None`

Returns:

Type	Description
`DatasetMetadata`	Metadata for the new (filtered) dataset.

bandstop

bandstop(
    dataset, low_freq, high_freq, output_path, *, order=None
)

Apply a band-stop (band-reject) filter to a dataset.

Parameters:

Name	Type	Description	Default
`dataset`	`FlexDataset`	Dataset to filter (metadata or ID).	required
`low_freq`	`float`	Lower cutoff frequency in Hz.	required
`high_freq`	`float`	Upper cutoff frequency in Hz.	required
`output_path`	`str \| Path`	Where to save the result.	required
`order`	`int \| None`	Filter order. If `None`, the server default is used.	`None`

Returns:

Type	Description
`DatasetMetadata`	Metadata for the new (filtered) dataset.

bulk_bandpass

bulk_bandpass(
    sources,
    low_freq,
    high_freq,
    output_dir,
    output_suffix,
    *,
    order=None,
)

Apply band-pass filter to multiple datasets at once.

Parameters:

Name	Type	Description	Default
`sources`	`Sequence[FlexDataset]`	Datasets to filter (metadata or IDs).	required
`low_freq`	`float`	Lower cutoff frequency in Hz.	required
`high_freq`	`float`	Upper cutoff frequency in Hz.	required
`output_dir`	`str \| Path`	Directory where output files will be written.	required
`output_suffix`	`str`	Suffix appended to each output file's base name.	required
`order`	`int \| None`	Filter order. If `None`, the server default is used.	`None`

bulk_bandstop

bulk_bandstop(
    sources,
    low_freq,
    high_freq,
    output_dir,
    output_suffix,
    *,
    order=None,
)

Apply band-stop filter to multiple datasets at once.

Parameters:

Name	Type	Description	Default
`sources`	`Sequence[FlexDataset]`	Datasets to filter (metadata or IDs).	required
`low_freq`	`float`	Lower cutoff frequency in Hz.	required
`high_freq`	`float`	Upper cutoff frequency in Hz.	required
`output_dir`	`str \| Path`	Directory where output files will be written.	required
`output_suffix`	`str`	Suffix appended to each output file's base name.	required
`order`	`int \| None`	Filter order. If `None`, the server default is used.	`None`

bulk_downsample

bulk_downsample(sources, factor, output_dir, output_suffix)

Downsample multiple datasets at once.

Parameters:

Name	Type	Description	Default
`sources`	`Sequence[FlexDataset]`	Datasets to downsample (metadata or IDs).	required
`factor`	`int`	Downsampling factor (e.g., 2, 4, 10).	required
`output_dir`	`str \| Path`	Directory where output files will be written.	required
`output_suffix`	`str`	Suffix appended to each output file's base name.	required

bulk_highpass

bulk_highpass(
    sources,
    frequency,
    output_dir,
    output_suffix,
    *,
    order=None,
)

Apply high-pass filter to multiple datasets at once.

Parameters:

Name	Type	Description	Default
`sources`	`Sequence[FlexDataset]`	Datasets to filter (metadata or IDs).	required
`frequency`	`float`	Cutoff frequency in Hz.	required
`output_dir`	`str \| Path`	Directory where output files will be written.	required
`output_suffix`	`str`	Suffix appended to each output file's base name.	required
`order`	`int \| None`	Filter order. If `None`, the server default is used.	`None`

bulk_lowpass

bulk_lowpass(
    sources,
    frequency,
    output_dir,
    output_suffix,
    *,
    order=None,
)

Apply low-pass filter to multiple datasets at once.

Parameters:

Name	Type	Description	Default
`sources`	`Sequence[FlexDataset]`	Datasets to filter (metadata or IDs).	required
`frequency`	`float`	Cutoff frequency in Hz.	required
`output_dir`	`str \| Path`	Directory where output files will be written.	required
`output_suffix`	`str`	Suffix appended to each output file's base name.	required
`order`	`int \| None`	Filter order. If `None`, the server default is used.	`None`

bulk_notch

bulk_notch(
    sources,
    notch_freq,
    bandwidth,
    output_dir,
    output_suffix,
)

Apply notch filter to multiple datasets at once.

The notch filter is always second-order (scipy.signal.iirnotch). Unlike other IIR filters, it does not accept a custom order.

Parameters:

Name	Type	Description	Default
`sources`	`Sequence[FlexDataset]`	Datasets to filter (metadata or IDs).	required
`notch_freq`	`float`	Center frequency to reject in Hz.	required
`bandwidth`	`float`	Width of the rejection band in Hz.	required
`output_dir`	`str \| Path`	Directory where output files will be written.	required
`output_suffix`	`str`	Suffix appended to each output file's base name.	required

bulk_slice_time

bulk_slice_time(
    sources, start_sec, end_sec, output_dir, output_suffix
)

Slice multiple datasets in time at once.

Parameters:

Name	Type	Description	Default
`sources`	`Sequence[FlexDataset]`	Datasets to slice (metadata or IDs).	required
`start_sec`	`float`	Start time in seconds.	required
`end_sec`	`float`	End time in seconds.	required
`output_dir`	`str \| Path`	Directory where output files will be written.	required
`output_suffix`	`str`	Suffix appended to each output file's base name.	required

car

car(dataset, output_path)

Apply Common Average Reference (CAR) re-referencing to a dataset.

Subtracts the mean signal across all channels from each channel. This is the most widely used reference scheme in LFP and spike analysis.

Parameters:

Name	Type	Description	Default
`dataset`	`FlexDataset`	Dataset to re-reference (metadata or ID).	required
`output_path`	`str \| Path`	Where to save the result.	required

Returns:

Type	Description
`DatasetMetadata`	Metadata for the new (re-referenced) dataset.

clear_data_sources

clear_data_sources(dsource_ids)

Clear (unlink) data sources from the server.

Parameters:

Name	Type	Description	Default
`dsource_ids`	`list[str] \| str`	List of IDs (or a single ID) to clear.	required

Returns:

Type	Description
`list[str]`	List of successfully cleared IDs.

close

close()

Close all gRPC channels managed by this client.

copy_data_source_file

copy_data_source_file(src_path, dest_path, *, force=False)

Copy a server-side data source file to a new location.

Parameters:

Name	Type	Description	Default
`src_path`	`str \| Path`	Source file path on the server.	required
`dest_path`	`str \| Path`	Destination file path on the server.	required
`force`	`bool`	If `True`, overwrite existing destination file.	`False`

downsample

downsample(dataset, factor, output_path)

Downsample a dataset.

Parameters:

Name	Type	Description	Default
`dataset`	`FlexDataset`	Dataset to downsample (metadata or ID).	required
`factor`	`int`	Downsampling factor (e.g., 2, 4, 10).	required
`output_path`	`str \| Path`	Where to save the result.	required

Returns:

Type	Description
`DatasetMetadata`	Metadata for the new (downsampled) dataset.

export_data_source

export_data_source(dsource, dest_path, force=False)

Export a data source to a new file.

Parameters:

Name	Type	Description	Default
`dsource`	`FlexDataset`	The data source to export (ID or metadata object).	required
`dest_path`	`str \| Path`	Destination file path.	required
`force`	`bool`	If `True`, overwrite existing destination file.	`False`

get_all_protocols

get_all_protocols()

List all saved protocols on the server.

Returns:

Type	Description
`list[ProtocolSpec]`	Summary of every protocol currently registered.

get_protocol

get_protocol(protocol_id)

Retrieve a saved protocol by ID.

Parameters:

Name	Type	Description	Default
`protocol_id`	`str`	ID of the protocol to retrieve.	required

Returns:

Type	Description
`ProtocolSpec`	Minimal summary of the protocol.

healthcheck

healthcheck(service=ServiceEndpoint.RADIENS_LIFECYCLE)

Verify the radiensserver is responsive.

Parameters:

Name	Type	Description	Default
`service`	`ServiceEndpoint`	Which gRPC service to probe (default `RADIENS_LIFECYCLE`). `RADIENS_CORE` is also supported.	`RADIENS_LIFECYCLE`

Raises:

Type	Description
`RadiensError`	If the health check fails or the server is unresponsive.

highpass

highpass(dataset, frequency, output_path, *, order=None)

Apply a high-pass filter to a dataset.

Parameters:

Name	Type	Description	Default
`dataset`	`FlexDataset`	Dataset to filter (metadata or ID).	required
`frequency`	`float`	Cutoff frequency in Hz.	required
`output_path`	`str \| Path`	Where to save the result.	required
`order`	`int \| None`	Filter order. If `None`, the server default is used.	`None`

Returns:

Type	Description
`DatasetMetadata`	Metadata for the new (filtered) dataset.

link_data_file

link_data_file(fpath, file_type=None)

Link a recording file and return its metadata.

Also discovers and registers any associated .spikes file found alongside the recording. If found, the returned DatasetMetadata will have associated_spikes_ids populated, enabling the spike data methods on VidereClient.

Parameters:

Name	Type	Description	Default
`path`		Path to the recording file (e.g. `.xdat`, `.rhd`).	required

Returns:

Type	Description
`DatasetMetadata`	Metadata for the linked dataset.

Raises:

Type	Description
`RadiensError`	On communication failure or server error.

list_data_source_ids

list_data_source_ids()

List IDs of all currently linked data sources.

Returns:

Type	Description
`list[str]`	List of data source IDs.

list_data_sources

list_data_sources(dsource_ids=None)

List linked data sources and their metadata.

Parameters:

Name	Type	Description	Default
`dsource_ids`	`list[str] \| str \| None`	Optional list of specific IDs to query. If `None`, lists all.	`None`

Returns:

Type	Description
`dict[str, DatasetMetadata]`	Map of data source ID to metadata.

list_directory

list_directory(directory)

List files in a remote server-side directory.

Parameters:

Name	Type	Description	Default
`directory`	`str \| Path`	Server-side directory path to list.	required

Returns:

Type	Description
`list[FileInfo]`	Metadata for each file found in the directory.

lowpass

lowpass(dataset, frequency, output_path, *, order=None)

Apply a low-pass filter to a dataset.

Parameters:

Name	Type	Description	Default
`dataset`	`FlexDataset`	Dataset to filter (metadata or ID).	required
`frequency`	`float`	Cutoff frequency in Hz.	required
`output_path`	`str \| Path`	Where to save the result.	required
`order`	`int \| None`	Filter order. If `None`, the server default is used.	`None`

Returns:

Type	Description
`DatasetMetadata`	Metadata for the new (filtered) dataset.

move_data_source_file

move_data_source_file(src_path, dest_path, *, force=False)

Move a server-side data source file to a new location.

Parameters:

Name	Type	Description	Default
`src_path`	`str \| Path`	Source file path on the server.	required
`dest_path`	`str \| Path`	Destination file path on the server.	required
`force`	`bool`	If `True`, overwrite existing destination file.	`False`

notch

notch(dataset, frequency, bandwidth, output_path)

Apply a notch filter to a dataset.

The notch filter is always second-order (scipy.signal.iirnotch). Unlike other IIR filters, it does not accept a custom order.

Parameters:

Name	Type	Description	Default
`dataset`	`FlexDataset`	Dataset to filter (metadata or ID).	required
`frequency`	`float`	Center frequency to reject, in Hz (e.g., 60.0 for line noise).	required
`bandwidth`	`float`	Width of the rejection band in Hz.	required
`output_path`	`str \| Path`	Where to save the result.	required

Returns:

Type	Description
`DatasetMetadata`	Metadata for the new (filtered) dataset.

paired_ref

paired_ref(
    dataset, ref_channel, target_channel, output_path
)

Apply paired re-referencing between two specific channels.

References one target channel against one reference channel.

Parameters:

Name	Type	Description	Default
`dataset`	`FlexDataset`	Dataset to re-reference (metadata or ID).	required
`ref_channel`	`int`	Native channel index of the reference electrode.	required
`target_channel`	`int`	Native channel index of the channel to re-reference.	required
`output_path`	`str \| Path`	Where to save the result.	required

Returns:

Type	Description
`DatasetMetadata`	Metadata for the new (re-referenced) dataset.

remove_data_source_file

remove_data_source_file(path, *, force=False)

Remove a server-side data source file.

Parameters:

Name	Type	Description	Default
`path`	`str \| Path`	File path on the server to remove.	required
`force`	`bool`	If `True`, force removal even if the file is in use.	`False`

set_protocol

set_protocol(nodes, edges, *, protocol_id=None)

Register a custom protocol with the server.

Build a directed acyclic graph of transform nodes and register it for later execution with apply_protocol(). The server validates the graph and returns a summary.

Use TransformNode factory methods to construct nodes, and TransformEdge to wire them together::

src  = TransformNode.source(meta)
hp   = TransformNode.highpass(300.0)
car  = TransformNode.car()
snk  = TransformNode.sink("result", "/output", RadiensFileType.XDAT)

edges = [
    TransformEdge(source=src.id, target=hp.id),
    TransformEdge(source=hp.id, target=car.id),
    TransformEdge(source=car.id, target=snk.id),
]

spec = client.set_protocol([src, hp, car, snk], edges)
client.apply_protocol(spec.id)

Parameters:

Name	Type	Description	Default
`nodes`	`list[TransformNode]`	Ordered list of transform nodes forming the graph.	required
`edges`	`list[TransformEdge]`	Directed edges connecting the nodes by ID.	required
`protocol_id`	`str \| None`	Optional protocol ID. A UUID is generated if not provided.	`None`

Returns:

Type	Description
`ProtocolSpec`	Summary of the validated (server-registered) protocol.

Raises:

Type	Description
`RadiensError`	If the server marks one or more nodes as invalid.

slice_channels

slice_channels(dataset, channels, output_path)

Select a subset of channels from a dataset.

Parameters:

Name	Type	Description	Default
`dataset`	`FlexDataset`	Dataset to slice (metadata or ID).	required
`channels`	`Sequence[int]`	System-wide channel indices to keep.	required
`output_path`	`str \| Path`	Where to save the result.	required

Returns:

Type	Description
`DatasetMetadata`	Metadata for the new dataset containing only the selected channels.

slice_time

slice_time(dataset, start_sec, end_sec, output_path)

Slice a dataset in time.

Parameters:

Name	Type	Description	Default
`dataset`	`FlexDataset`	Dataset to slice (metadata or ID).	required
`start_sec`	`float`	Start time in seconds.	required
`end_sec`	`float`	End time in seconds.	required
`output_path`	`str \| Path`	Where to save the result.	required

Returns:

Type	Description
`DatasetMetadata`	Metadata for the new (sliced) dataset.

virtual_ref

virtual_ref(dataset, ref_channel, output_path)

Apply virtual reference re-referencing to a dataset.

References all channels against a single nominated electrode.

Parameters:

Name	Type	Description	Default
`dataset`	`FlexDataset`	Dataset to re-reference (metadata or ID).	required
`ref_channel`	`int`	Native channel index of the reference electrode.	required
`output_path`	`str \| Path`	Where to save the result.	required

Returns:

Type	Description
`DatasetMetadata`	Metadata for the new (re-referenced) dataset.