Data I/O Blocks
Blocks for loading data from disk and saving results. General-purpose blocks work with any array format; CRCNS I/O blocks are purpose-built for the HC-11 dataset.
Load .npy file
load_npy
Loads a NumPy .npy or .npz file from disk and wraps it in a NeuroData envelope.
Ports
| Port | Direction | Type | Description |
|---|---|---|---|
data |
Output | NeuroData[raw_signal] |
Loaded array |
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
file_path |
str | "" |
Absolute path to the .npy or .npz file |
sampling_rate |
float | 1000.0 |
Samples per second |
data_type_override |
enum | raw_signal |
Override the NeuroData type tag: raw_signal, lfp, spike_times, or position |
Load CSV
load_csv
Loads a CSV file. If has_timestamps is true the first column is treated as timestamps (seconds) and excluded from the data array.
Ports
| Port | Direction | Type | Description |
|---|---|---|---|
data |
Output | NeuroData[raw_signal] |
Loaded array |
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
file_path |
str | "" |
Absolute path to the CSV file |
has_timestamps |
bool | true |
If true, first column is timestamps (seconds) |
sampling_rate |
float | 1000.0 |
Sampling rate used when has_timestamps is false |
Save .npy file
save_npy
Saves a NeuroData array to disk as a .npy file.
Ports
| Port | Direction | Type | Description |
|---|---|---|---|
data |
Input | NeuroData[any] |
Data to save |
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
file_path |
str | "" |
Destination path (e.g. /data/out.npy) |
overwrite |
bool | false |
If false, raises an error if the file already exists |
Export CSV
export_csv
Exports a NeuroData array to a CSV file. Optionally prepends a timestamps column.
Ports
| Port | Direction | Type | Description |
|---|---|---|---|
data |
Input | NeuroData[any] |
Data to export |
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
file_path |
str | "" |
Destination path (e.g. /data/out.csv) |
include_timestamps |
bool | true |
If true, prepend the timestamps column |
Load CRCNS Session
load_crcns_session
Reads *_sessInfo.mat from a CRCNS HC-11 session folder and returns multi-cell spike times, linearised 1-D position, and epoch boundaries. Requires the HDF5-format .mat file (MATLAB v7.3).
Ports
| Port | Direction | Type | Description |
|---|---|---|---|
spike_data |
Output | NeuroData[multi_spike_times] |
All spike times with per-spike cell IDs in metadata |
position |
Output | NeuroData[position] |
Linearised 1-D position (cm) at ~39 Hz, maze epoch only |
epochs_info |
Output | NeuroData[epochs] |
Epoch boundaries: MazeEpoch, PREEpoch, POSTEpoch, Wake, NREM, REM |
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
session_path |
str | "" |
Absolute path to the session folder (e.g. .../Achilles_10252013) |
cell_type_filter |
enum | all |
pyramidal = PyrIDs only; interneuron = IntIDs only; all = both |
track_length_cm |
float | 160.0 |
Known track length in cm; raw position range is scaled to [0, track_length_cm] |
Load CRCNS LFP
load_crcns_lfp
Loads selected channels from a CRCNS HC-11 baseName.eeg file (int16, 1250 Hz). Connect a NeuroData[epochs] time range to avoid reading the full ~11 GB file.
Ports
| Port | Direction | Type | Description |
|---|---|---|---|
time_range |
Input (optional) | NeuroData[epochs] |
Epoch window from select_epoch; overrides t_start/t_end parameters |
lfp |
Output | NeuroData[lfp] |
LFP signal (n_samples × n_channels) at 1250 Hz |
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
session_path |
str | "" |
Path to the session folder (same folder as the .eeg file) |
channels |
str | "0" |
Comma-separated channel indices or range, e.g. "0,1,2" or "0-9" |
t_start |
float | -1.0 |
Start time in seconds; -1 = beginning of file |
t_end |
float | -1.0 |
End time in seconds; -1 = end of file |
invert_signal |
bool | false |
Multiply signal by −1 (for recordings with inverted polarity) |