Skip to content

Add more options to fit_psp() #6

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
meganbkratz wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Add more options to fit_psp() #6

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 11 additions & 3 deletions neuroanalysis/fitting/psp.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -172,7 +172,7 @@ def double_psp_func(x, xoffset, yoffset, rise_tau, decay_tau1, decay_tau2, amp1,
return out


def fit_psp(data, search_window, clamp_mode, sign=0, exp_baseline=True, baseline_like_psp=False, refine=True, init_params=None, decay_tau_bounds=None, rise_time_bounds=None, fit_kws=None, ui=None):
def fit_psp(data, search_window, clamp_mode, sign=0, exp_baseline=True, baseline_like_psp=False, refine=True, init_params=None, decay_tau_bounds=None, rise_time_bounds=None, fit_kws=None, ui=None, fine_search_spacing=0.2e-3, fine_search_window_width=1e-3):
"""Fit a Trace instance to a StackedPsp model.

This function is a higher-level interface to StackedPsp.fit:
Expand Down Expand Up @@ -208,6 +208,12 @@ def fit_psp(data, search_window, clamp_mode, sign=0, exp_baseline=True, baseline
(min, max) values for rise_time. Default bounds are the initial value for rise_time 0.1 and 10.
fit_kws : dict
Extra keyword arguments to send to the minimizer
ui : ? | None
fine_search_spacing: float | 0.2 ms
How finely to divide the xoffset parameter space for the fine search, only relevant if refine=True
fine_search_window_width: float | 1 ms
Width of the window for the fine search. Window will be the xoffset result of the
course search +/- fine_search_window_width. Only relevant if refine=True

Returns
-------
Expand Down Expand Up @@ -359,14 +365,15 @@ def fit_psp(data, search_window, clamp_mode, sign=0, exp_baseline=True, baseline
# Round 2: fine fit

# Fine search xoffset
fine_search_window = (max(search_window[0], fit['xoffset']-1e-3), min(search_window[1], fit['xoffset']+1e-3))
n_xoffset_chunks = max(1, int((fine_search_window[1] - fine_search_window[0]) / .2e-3))
fine_search_window = (max(search_window[0], fit['xoffset']-fine_search_window_width), min(search_window[1], fit['xoffset']+fine_search_window_width))
n_xoffset_chunks = max(1, int((fine_search_window[1] - fine_search_window[0]) / fine_search_spacing))
xoffset_chunks = np.linspace(fine_search_window[0], fine_search_window[1], n_xoffset_chunks + 1)
xoffset = [{'xoffset': ((a+b)/2., a, b)} for a,b in zip(xoffset_chunks[:-1], xoffset_chunks[1:])]

# Search amp / rise time / decay tau to avoid traps
rise_time_inits = base_params['rise_time'][0] * 1.2**np.arange(-3,6)
rise_time = [{'rise_time': (x,) + base_params['rise_time'][1:]} for x in rise_time_inits]
#print(' rise_times:', rise_time)

decay_tau_inits = base_params['decay_tau'][0] * 2.0**np.arange(-3,6)
decay_tau = [{'decay_tau': (x,) + base_params['decay_tau'][1:]} for x in decay_tau_inits]
Expand All @@ -388,6 +395,7 @@ def fit_psp(data, search_window, clamp_mode, sign=0, exp_baseline=True, baseline
search_params.append(amp)

prof("prepare for fine fit %r" % base_params)
#print(" total search params:", ['%s:%i'%(list(p[0].keys())[0], len(p)) for p in search_params])

# Find best fit
with warnings.catch_warnings():
Expand Down