Package PyDSTool :: Package Toolbox :: Module neuro_data
[hide private]
[frames] | no frames]

Module neuro_data

source code

Classes [hide private]
  get_spike_model
Qualitative test for presence of spike in model trajectory data using events to identify spike times.
  get_spike_data
Qualitative test for presence of spike in noisy data.
  get_burst_duration
  get_burst_active_phase
  get_burst_dc_offset
  get_burst_passive_extent
  burst_feature
Embed the following sub-features, if desired: get_burst_X, where X is a number of feature types defined in this module.
  get_burst_spikes
Requires a get_spike_data and get_spike_model instance to be the only sub-features (supplied as a dict with keys 'is_spike_data' and 'is_spike_model').
  get_burst_peak_env
Requires tol and num_samples parameters.
  get_burst_trough_env
Requires tol and num_samples parameters.
  get_burst_isi_env
Requires tol and num_samples parameters.
  get_burst_upsweep
  get_burst_downsweep
  get_burst_num_spikes
  get_burst_period_info
  spike_metric
Measures the distance between spike time and height, using an inherent weighting of height suited to neural voltage signals (0.05 of time distance).
  spike_feature
pars keys: tol
  geom_feature
Measures the residual between two 1D parameterized geometric curves (given as Trajectory objects).
  estimate_spiking
Estimate pattern of spiking in tonic or burst patterns.
  spike_envelope
Find an amplitude envelope over a smooth 1D signal that features roughly periodic spikes.
Functions [hide private]
 
find_internal_extrema(pts, noise_tol=0)
Find an interior (local) maximum and minimum values of a 1D pointset, away from the endpoints.
source code
Function Details [hide private]

find_internal_extrema(pts, noise_tol=0)

source code 

Find an interior (local) maximum and minimum values of a 1D pointset, away from the endpoints. Returns a dictionary mapping 'local_max' -> (index_max, xmax), 'local_min' -> (index_min, xmin), whose values are None if the pointset is monotonic or is close enough so that the global extrema are at the endpoints.

Use noise_tol > 0 to avoid getting a local extremum right next to an endpoint because of noise.

Also returned in the dictionary for reference: 'first' -> (0, <start_endpoint_value>), 'last' -> (last_index, <last_endpoint_value>), 'global_max' -> (index, value), 'global_min' -> (index, value)

Assumes there is only one interior (max, min) pair in pts, otherwise will return an arbitrary choice from multiple maxima and minima.