Skip to content

[FIX] Bounds incorrect in _drop_peak_overlap. #171

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

Merged
merged 2 commits into from
Apr 27, 2020
Merged

[FIX] Bounds incorrect in _drop_peak_overlap. #171

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
4ac7dd4
fixed bounds in _drop_peak_overlap
ryanhammonds Apr 18, 2020
824eb68
updated _gauss_overlap_thresh to 0.75
ryanhammonds Apr 21, 2020
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: 9 additions & 5 deletions fooof/objs/fit.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -177,7 +177,7 @@ def __init__(self, peak_width_limits=(0.5, 12.0), max_n_peaks=np.inf, min_peak_h
self._bw_std_edge = 1.0
# Degree of overlap between gaussians for one to be dropped
# This is defined in units of gaussian standard deviation
self._gauss_overlap_thresh = 1.5
self._gauss_overlap_thresh = 0.75
# Parameter bounds for center frequency when fitting gaussians, in terms of +/- std dev
self._cf_bound = 1.5
# The maximum number of calls to the curve fitting function
Expand Down Expand Up @@ -816,7 +816,7 @@ def _fit_peaks(self, flat_iter):
max_ind = np.argmax(flat_iter)
max_height = flat_iter[max_ind]

# Stop searching for peaks peaks once drops below height threshold
# Stop searching for peaks once height drops below height threshold
if max_height <= self.peak_threshold * np.std(flat_iter):
break

Expand Down Expand Up @@ -1029,14 +1029,18 @@ def _drop_peak_overlap(self, guess):
the lowest height guess guassian is dropped.
"""

# Sort the peak guesses, so can check overlap of adjacent peaks
# Sort the peak guesses by increasing frequency, so adjacenent peaks can
# be compared from right to left.
guess = sorted(guess, key=lambda x: float(x[0]))

# Calculate standard deviation bounds for checking amount of overlap
bounds = [[peak[0] - peak[2] * self._gauss_overlap_thresh, peak[0],
# The bounds are the gaussian frequncy +/- gaussian standard deviation
bounds = [[peak[0] - peak[2] * self._gauss_overlap_thresh,
peak[0] + peak[2] * self._gauss_overlap_thresh] for peak in guess]

# Loop through peak bounds, comparing current bound to that of next peak
# If the left peak's upper bound extends pass the right peaks lower bound,
# Then drop the guassian with the lower height.
drop_inds = []
for ind, b_0 in enumerate(bounds[:-1]):
b_1 = bounds[ind + 1]
Expand Down Expand Up @@ -1164,7 +1168,7 @@ def _prepare_data(freqs, power_spectrum, freq_range, spectra_dim=1, verbose=True
freqs, power_spectrum = trim_spectrum(freqs, power_spectrum, freq_range)

# Check if freqs start at 0 and move up one value if so
# Aperiodic fit gets an inf is freq of 0 is included, which leads to an error
# Aperiodic fit gets an inf if freq of 0 is included, which leads to an error
if freqs[0] == 0.0:
freqs, power_spectrum = trim_spectrum(freqs, power_spectrum, [freqs[1], freqs.max()])
if verbose:
Expand Down