def demo(parallel=False):
# roughly number of cores on your machine minus 1
n_processes = np.maximum(psutil.cpu_count() - 2, 1)
p = 2 # order of the AR model (in general 1 or 2)
stop_server()
if parallel:
start_server()
c = Client()
# LOAD MOVIE AND MEMORYMAP
fname_new = cm.save_memmap(['example_movies/demoMovie.tif'],
base_name='Yr')
Yr, dims, T = cm.load_memmap(fname_new)
# INIT
cnm = cnmf.CNMF(n_processes,
method_init='greedy_roi',
k=30,
gSig=[4, 4],
merge_thresh=.8,
p=p,
dview=c[:] if parallel else None,
Ain=None,
method_deconvolution='oasis')
# FIT
images = np.reshape(Yr.T, [T] + list(dims), order='F')
cnm = cnm.fit(images)
if parallel:
stop_server()
# verifying the spatial components
npt.assert_allclose(cnm.A.sum(), 32282000, 1e-3)
# verifying the temporal components
npt.assert_allclose(cnm.C.sum(), 640.5, 1e-2)
def main(n_processes=None, patches=True, rf=64):
t = -time()
Yr, dims, T = cm.load_memmap(os.path.abspath('./Yr_d1_253_d2_316_d3_1_order_C_frames_2000_.mmap'))
Y = Yr.T.reshape((T,) + dims, order='F')
# c, dview, n_processes = cm.cluster.setup_cluster(backend='local', n_processes=n_processes)
# above line doesn't work cause memory_profiler creates some multiprocessing object itself
if n_processes is None:
n_processes = cpu_count()
dview = Pool(n_processes) if patches else None
print('{0} processes'.format(n_processes))
patch_args = dict(nb_patch=0, del_duplicates=True, rf=(rf, rf), stride=(16, 16)) \
if patches else {}
cnm = cnmf.CNMF(n_processes=n_processes, method_init='corr_pnr', k=None, dview=dview,
gSig=(3, 3), gSiz=(10, 10), merge_thresh=.8, p=1, tsub=2, ssub=1,
only_init_patch=True, gnb=0, min_corr=.9, min_pnr=15, normalize_init=False,
ring_size_factor=1.5, center_psf=True, ssub_B=2, init_iter=1, **patch_args)
cnm.fit(Y)
if patches:
dview.terminate()
t += time()
sleep(1) # just in case Pool takes some time to terminate
return t
def fit_cnmfe(self):
"""
Do CNMF-E
"""
self.cnm = cnmf.CNMF(n_processes=self.n_processes,
dview=self.dview,
Ain=None,
params=self.opts)
self.cnm.fit(self.images)
def demo(parallel=False):
p = 2 # order of the AR model (in general 1 or 2)
if parallel:
c, dview, n_processes = cm.cluster.setup_cluster(backend='local',
n_processes=None,
single_thread=False)
else:
n_processes, dview = 2, None
# LOAD MOVIE AND MEMORYMAP
fname_new = cm.save_memmap(
[os.path.join(caiman_datadir(), 'example_movies', 'demoMovie.tif')],
base_name='Yr',
order='C')
Yr, dims, T = cm.load_memmap(fname_new)
# INIT
cnm = cnmf.CNMF(n_processes,
method_init='greedy_roi',
k=30,
gSig=[4, 4],
merge_thresh=.8,
p=p,
dview=dview,
Ain=None,
method_deconvolution='oasis')
# FIT
images = np.reshape(Yr.T, [T] + list(dims), order='F')
cnm = cnm.fit(images)
if parallel:
cm.cluster.stop_server(dview=dview)
# verifying the spatial components
npt.assert_allclose(cnm.estimates.A.sum(), 281.1, 1e-2)
# verifying the temporal components
npt.assert_allclose(cnm.estimates.C.sum(), 66271668, 1e-2)
try:
dview.terminate()
except:
pass
def pipeline(D):
#%% GENERATE GROUND TRUTH DATA
Yr, trueC, trueS, trueA, centers, dims = gen_data(D)
N, T = trueC.shape
# INIT
params = caiman.source_extraction.cnmf.params.CNMFParams(
dims=dims,
k=4,
gSig=[2, 2, 2][:D],
p=1,
n_pixels_per_process=np.prod(dims),
block_size_spat=np.prod(dims),
block_size_temp=np.prod(dims))
params.spatial['thr_method'] = 'nrg'
params.spatial['extract_cc'] = False
cnm = cnmf.CNMF(2, params=params)
# FIT
images = np.reshape(Yr.T, (T, ) + dims, order='F')
cnm = cnm.fit(images)
# VERIFY HIGH CORRELATION WITH GROUND TRUTH
sorting = [
np.argmax([np.corrcoef(tc, c)[0, 1] for tc in trueC])
for c in cnm.estimates.C
]
# verifying the temporal components
corr = [
np.corrcoef(trueC[sorting[i]], cnm.estimates.C[i])[0, 1]
for i in range(N)
]
npt.assert_allclose(corr, 1, .05)
# verifying the spatial components
corr = [
np.corrcoef(
np.reshape(trueA, (-1, 4), order='F')[:, sorting[i]],
cnm.estimates.A.toarray()[:, i])[0, 1] for i in range(N)
]
npt.assert_allclose(corr, 1, .05)
def demo(parallel=False):
p = 2 # order of the AR model (in general 1 or 2)
if parallel:
c, dview, n_processes = cm.cluster.setup_cluster(backend='local',
n_processes=None,
single_thread=False)
else:
n_processes, dview = 2, None
# LOAD MOVIE AND MEMORYMAP
fname_new = cm.save_memmap([
os.path.abspath(cm.__path__[0][:-7]) + '/example_movies/demoMovie.tif'
],
base_name='Yr')
Yr, dims, T = cm.load_memmap(fname_new)
# INIT
cnm = cnmf.CNMF(n_processes,
method_init='greedy_roi',
k=30,
gSig=[4, 4],
merge_thresh=.8,
p=p,
dview=dview,
Ain=None,
method_deconvolution='oasis')
# FIT
images = np.reshape(Yr.T, [T] + list(dims), order='F')
cnm = cnm.fit(images)
if parallel:
cm.cluster.stop_server()
# verifying the spatial components
npt.assert_allclose(cnm.A.sum(), 31913160, 1e-2)
# verifying the temporal components
npt.assert_allclose(cnm.C.sum(), 640.5, 1e-2)
def main():
pass # For compatibility between running under Spyder and the CLI
#%% First setup some parameters
# dataset dependent parameters
display_images = False # Set to true to show movies and images
fnames = ['data_endoscope.tif'] # filename to be processed
frate = 10 # movie frame rate
decay_time = 0.4 # length of a typical transient in seconds
# motion correction parameters
do_motion_correction_nonrigid = True
do_motion_correction_rigid = False # in this case it will also save a rigid motion corrected movie
gSig_filt = (3, 3) # size of filter, in general gSig (see below),
# change this one if algorithm does not work
max_shifts = (5, 5) # maximum allowed rigid shift
splits_rig = 10 # for parallelization split the movies in num_splits chuncks across time
strides = (
48, 48
) # start a new patch for pw-rigid motion correction every x pixels
overlaps = (24, 24
) # overlap between pathes (size of patch strides+overlaps)
# for parallelization split the movies in num_splits chuncks across time
# (remember that it should hold that length_movie/num_splits_to_process_rig>100)
splits_els = 10
upsample_factor_grid = 4 # upsample factor to avoid smearing when merging patches
# maximum deviation allowed for patch with respect to rigid shifts
max_deviation_rigid = 3
# parameters for source extraction and deconvolution
p = 1 # order of the autoregressive system
K = None # upper bound on number of components per patch, in general None
gSig = 3 # gaussian width of a 2D gaussian kernel, which approximates a neuron
gSiz = 13 # average diameter of a neuron, in general 4*gSig+1
merge_thresh = .7 # merging threshold, max correlation allowed
rf = 40 # half-size of the patches in pixels. e.g., if rf=40, patches are 80x80
stride_cnmf = 20 # amount of overlap between the patches in pixels
# (keep it at least large as gSiz, i.e 4 times the neuron size gSig)
tsub = 2 # downsampling factor in time for initialization,
# increase if you have memory problems
ssub = 1 # downsampling factor in space for initialization,
# increase if you have memory problems
Ain = None # if you want to initialize with some preselected components
# you can pass them here as boolean vectors
low_rank_background = None # None leaves background of each patch intact,
# True performs global low-rank approximation
gnb = -1 # number of background components (rank) if positive,
# else exact ring model with following settings
# gnb=-2: Return background as b and W
# gnb=-1: Return full rank background B
# gnb= 0: Don't return background
nb_patch = -1 # number of background components (rank) per patch,
# use 0 or -1 for exact background of ring model (cf. gnb)
min_corr = .8 # min peak value from correlation image
min_pnr = 10 # min peak to noise ration from PNR image
ssub_B = 2 # additional downsampling factor in space for background
ring_size_factor = 1.4 # radius of ring is gSiz*ring_size_factor
# parameters for component evaluation
min_SNR = 3 # adaptive way to set threshold on the transient size
r_values_min = 0.85 # threshold on space consistency (if you lower more components
# will be accepted, potentially with worst quality)
#%% start the cluster
try:
cm.stop_server(dview=dview) # stop it if it was running
except:
pass
c, dview, n_processes = cm.cluster.setup_cluster(
backend='local', # use this one
n_processes=
24, # number of process to use, if you go out of memory try to reduce this one
single_thread=False)
#%% download demo file
fnames = [download_demo(fnames[0])]
filename_reorder = fnames
#%% MOTION CORRECTION
if do_motion_correction_nonrigid or do_motion_correction_rigid:
# do motion correction rigid
mc = motion_correct_oneP_rigid(
fnames,
gSig_filt=gSig_filt,
max_shifts=max_shifts,
dview=dview,
splits_rig=splits_rig,
save_movie=not (do_motion_correction_nonrigid),
border_nan='copy')
new_templ = mc.total_template_rig
plt.subplot(1, 2, 1)
plt.imshow(new_templ) # % plot template
plt.subplot(1, 2, 2)
plt.plot(mc.shifts_rig) # % plot rigid shifts
plt.legend(['x shifts', 'y shifts'])
plt.xlabel('frames')
plt.ylabel('pixels')
# borders to eliminate from movie because of motion correction
bord_px = np.ceil(np.max(np.abs(mc.shifts_rig))).astype(np.int)
filename_reorder = mc.fname_tot_rig
# do motion correction nonrigid
if do_motion_correction_nonrigid:
mc = motion_correct_oneP_nonrigid(
fnames,
gSig_filt=gSig_filt,
max_shifts=max_shifts,
strides=strides,
overlaps=overlaps,
splits_els=splits_els,
upsample_factor_grid=upsample_factor_grid,
max_deviation_rigid=max_deviation_rigid,
dview=dview,
splits_rig=None,
save_movie=True, # whether to save movie in memory mapped format
new_templ=new_templ, # template to initialize motion correction
border_nan='copy')
filename_reorder = mc.fname_tot_els
bord_px = np.ceil(
np.maximum(np.max(np.abs(mc.x_shifts_els)),
np.max(np.abs(mc.y_shifts_els)))).astype(np.int)
# create memory mappable file in the right order on the hard drive (C order)
fname_new = cm.save_memmap(filename_reorder,
base_name='memmap_',
order='C',
border_to_0=bord_px,
dview=dview)
# load memory mappable file
Yr, dims, T = cm.load_memmap(fname_new)
Y = Yr.T.reshape((T, ) + dims, order='F')
#%% compute some summary images (correlation and peak to noise)
# change swap dim if output looks weird, it is a problem with tiffile
cn_filter, pnr = cm.summary_images.correlation_pnr(Y,
gSig=gSig,
swap_dim=False)
# inspect the summary images and set the parameters
inspect_correlation_pnr(cn_filter, pnr)
# print parameters set above, modify them if necessary based on summary images
print(min_corr) # min correlation of peak (from correlation image)
print(min_pnr) # min peak to noise ratio
#%% RUN CNMF ON PATCHES
cnm = cnmf.CNMF(
n_processes=n_processes,
method_init='corr_pnr', # use this for 1 photon
k=K,
gSig=(gSig, gSig),
gSiz=(gSiz, gSiz),
merge_thresh=merge_thresh,
p=p,
dview=dview,
tsub=tsub,
ssub=ssub,
Ain=Ain,
rf=rf,
stride=stride_cnmf,
only_init_patch=True, # just leave it as is
gnb=gnb,
nb_patch=nb_patch,
method_deconvolution='oasis', # could use 'cvxpy' alternatively
low_rank_background=low_rank_background,
update_background_components=
True, # sometimes setting to False improve the results
min_corr=min_corr,
min_pnr=min_pnr,
normalize_init=False, # just leave as is
center_psf=True, # leave as is for 1 photon
ssub_B=ssub_B,
ring_size_factor=ring_size_factor,
del_duplicates=True, # whether to remove duplicates from initialization
border_pix=bord_px) # number of pixels to not consider in the borders
cnm.fit(Y)
#%% DISCARD LOW QUALITY COMPONENTS
cnm.evaluate_components(Y,
min_SNR=min_SNR,
rval_thr=r_values_min,
use_cnn=False,
decay_time=decay_time,
fr=frate)
print(' ***** ')
print('Number of total components: ', len(cnm.C))
print('Number of accepted components: ', len(cnm.idx_components))
#%% PLOT COMPONENTS
cnm.dims = dims
if display_images:
cnm.plot_contours(img=cn_filter, idx=cnm.idx_components)
cnm.view_components(Y, dims, idx=cnm.idx_components)
#%% MOVIES
if display_images:
# fully reconstructed movie
cnm.play_movie(Y, magnification=3, include_bck=True, gain_res=10)
# movie without background
cnm.play_movie(Y, magnification=3, include_bck=False, gain_res=4)
#%% STOP SERVER
cm.stop_server(dview=dview)
def main():
pass # For compatibility between running under Spyder and the CLI
# %% start the cluster
try:
cm.stop_server() # stop it if it was running
except ():
pass
c, dview, n_processes = cm.cluster.setup_cluster(
backend='local',
n_processes=
24, # number of process to use, if you go out of memory try to reduce this one
single_thread=False)
# %% First setup some parameters for motion correction
# dataset dependent parameters
fnames = ['data_endoscope.tif'] # filename to be processed
fnames = [download_demo(fnames[0])] # download file if not already present
filename_reorder = fnames
fr = 10 # movie frame rate
decay_time = 0.4 # length of a typical transient in seconds
# motion correction parameters
motion_correct = True # flag for motion correction
pw_rigid = False # flag for pw-rigid motion correction
gSig_filt = (3, 3) # size of filter, in general gSig (see below),
# change this one if algorithm does not work
max_shifts = (5, 5) # maximum allowed rigid shift
strides = (
48, 48
) # start a new patch for pw-rigid motion correction every x pixels
overlaps = (24, 24
) # overlap between pathes (size of patch strides+overlaps)
# maximum deviation allowed for patch with respect to rigid shifts
max_deviation_rigid = 3
border_nan = 'copy'
mc_dict = {
'fnames': fnames,
'fr': fr,
'decay_time': decay_time,
'pw_rigid': pw_rigid,
'max_shifts': max_shifts,
'gSig_filt': gSig_filt,
'strides': strides,
'overlaps': overlaps,
'max_deviation_rigid': max_deviation_rigid,
'border_nan': border_nan
}
opts = params.CNMFParams(params_dict=mc_dict)
# %% MOTION CORRECTION
# The pw_rigid flag set above, determines where to use rigid or pw-rigid
# motion correction
if motion_correct:
# do motion correction rigid
mc = MotionCorrect(fnames, dview=dview, **opts.get_group('motion'))
mc.motion_correct(save_movie=True)
fname_mc = mc.fname_tot_els if pw_rigid else mc.fname_tot_rig
if pw_rigid:
bord_px = np.ceil(
np.maximum(np.max(np.abs(mc.x_shifts_els)),
np.max(np.abs(mc.y_shifts_els)))).astype(np.int)
else:
bord_px = np.ceil(np.max(np.abs(mc.shifts_rig))).astype(np.int)
plt.subplot(1, 2, 1)
plt.imshow(mc.total_template_rig) # % plot template
plt.subplot(1, 2, 2)
plt.plot(mc.shifts_rig) # % plot rigid shifts
plt.legend(['x shifts', 'y shifts'])
plt.xlabel('frames')
plt.ylabel('pixels')
bord_px = 0 if border_nan == 'copy' else bord_px
fname_new = cm.save_memmap(fname_mc,
base_name='memmap_',
order='C',
border_to_0=bord_px)
else: # if no motion correction just memory map the file
fname_new = cm.save_memmap(filename_reorder,
base_name='memmap_',
order='C',
border_to_0=0,
dview=dview)
# load memory mappable file
Yr, dims, T = cm.load_memmap(fname_new)
images = Yr.T.reshape((T, ) + dims, order='F')
# %% Parameters for source extraction and deconvolution (CNMF-E algorithm)
p = 1 # order of the autoregressive system
K = None # upper bound on number of components per patch, in general None for 1p data
gSig = (
3, 3
) # gaussian width of a 2D gaussian kernel, which approximates a neuron
gSiz = (13, 13) # average diameter of a neuron, in general 4*gSig+1
Ain = None # possibility to seed with predetermined binary masks
merge_thr = .7 # merging threshold, max correlation allowed
rf = 40 # half-size of the patches in pixels. e.g., if rf=40, patches are 80x80
stride_cnmf = 20 # amount of overlap between the patches in pixels
# (keep it at least large as gSiz, i.e 4 times the neuron size gSig)
tsub = 2 # downsampling factor in time for initialization,
# increase if you have memory problems
ssub = 1 # downsampling factor in space for initialization,
# increase if you have memory problems
# you can pass them here as boolean vectors
low_rank_background = None # None leaves background of each patch intact,
# True performs global low-rank approximation if gnb>0
gnb = 0 # number of background components (rank) if positive,
# else exact ring model with following settings
# gnb= 0: Return background as b and W
# gnb=-1: Return full rank background B
# gnb<-1: Don't return background
nb_patch = 0 # number of background components (rank) per patch if gnb>0,
# else it is set automatically
min_corr = .8 # min peak value from correlation image
min_pnr = 10 # min peak to noise ration from PNR image
ssub_B = 2 # additional downsampling factor in space for background
ring_size_factor = 1.4 # radius of ring is gSiz*ring_size_factor
opts.change_params(
params_dict={
'dims': dims,
'method_init': 'corr_pnr', # use this for 1 photon
'K': K,
'gSig': gSig,
'gSiz': gSiz,
'merge_thr': merge_thr,
'p': p,
'tsub': tsub,
'ssub': ssub,
'rf': rf,
'stride': stride_cnmf,
'only_init': True, # set it to True to run CNMF-E
'nb': gnb,
'nb_patch': nb_patch,
'method_deconvolution': 'oasis', # could use 'cvxpy' alternatively
'low_rank_background': low_rank_background,
'update_background_components':
True, # sometimes setting to False improve the results
'min_corr': min_corr,
'min_pnr': min_pnr,
'normalize_init': False, # just leave as is
'center_psf': True, # leave as is for 1 photon
'ssub_B': ssub_B,
'ring_size_factor': ring_size_factor,
'del_duplicates':
True, # whether to remove duplicates from initialization
'border_pix': bord_px
}) # number of pixels to not consider in the borders)
# %% compute some summary images (correlation and peak to noise)
# change swap dim if output looks weird, it is a problem with tiffile
cn_filter, pnr = cm.summary_images.correlation_pnr(images[::1],
gSig=gSig[0],
swap_dim=False)
# if your images file is too long this computation will take unnecessarily
# long time and consume a lot of memory. Consider changing images[::1] to
# images[::5] or something similar to compute on a subset of the data
# inspect the summary images and set the parameters
inspect_correlation_pnr(cn_filter, pnr)
# print parameters set above, modify them if necessary based on summary images
print(min_corr) # min correlation of peak (from correlation image)
print(min_pnr) # min peak to noise ratio
# %% RUN CNMF ON PATCHES
cnm = cnmf.CNMF(n_processes=n_processes, dview=dview, Ain=Ain, params=opts)
cnm.fit(images)
# %% ALTERNATE WAY TO RUN THE PIPELINE AT ONCE
# you can also perform the motion correction plus cnmf fitting steps
# simultaneously after defining your parameters object using
# cnm1 = cnmf.CNMF(n_processes, params=opts, dview=dview)
# cnm1.fit_file(motion_correct=True)
# %% DISCARD LOW QUALITY COMPONENTS
min_SNR = 2.5 # adaptive way to set threshold on the transient size
r_values_min = 0.85 # threshold on space consistency (if you lower more components
# will be accepted, potentially with worst quality)
cnm.params.set('quality', {
'min_SNR': min_SNR,
'rval_thr': r_values_min,
'use_cnn': False
})
cnm.estimates.evaluate_components(images, cnm.params, dview=dview)
print(' ***** ')
print('Number of total components: ', len(cnm.estimates.C))
print('Number of accepted components: ', len(cnm.estimates.idx_components))
# %% PLOT COMPONENTS
cnm.dims = dims
display_images = True # Set to true to show movies and images
if display_images:
cnm.estimates.plot_contours(img=cn_filter,
idx=cnm.estimates.idx_components)
cnm.estimates.view_components(images, idx=cnm.estimates.idx_components)
# %% MOVIES
display_images = False # Set to true to show movies and images
if display_images:
# fully reconstructed movie
cnm.estimates.play_movie(images,
q_max=99.5,
magnification=2,
include_bck=True,
gain_res=10,
bpx=bord_px)
# movie without background
cnm.estimates.play_movie(images,
q_max=99.9,
magnification=2,
include_bck=False,
gain_res=4,
bpx=bord_px)
# %% STOP SERVER
cm.stop_server(dview=dview)
def initialize_movie(Y,
K,
gSig,
rf,
stride,
base_name,
p=1,
merge_thresh=0.95,
rval_thr_online=0.9,
thresh_fitness_delta_online=-30,
thresh_fitness_raw_online=-50,
rval_thr_init=.5,
thresh_fitness_delta_init=-20,
thresh_fitness_raw_init=-20,
rval_thr_refine=0.95,
thresh_fitness_delta_refine=-100,
thresh_fitness_raw_refine=-100,
final_frate=10,
Npeaks=10,
single_thread=True,
dview=None):
_, d1, d2 = Y.shape
dims = (d1, d2)
Yr = Y.to_2D().T
# merging threshold, max correlation allowed
# order of the autoregressive system
#T = Y.shape[0]
base_name = base_name + '.mmap'
fname_new = Y.save(base_name, order='C')
#%
Yr, dims, T = cm.load_memmap(fname_new)
d1, d2 = dims
images = np.reshape(Yr.T, [T] + list(dims), order='F')
Y = np.reshape(Yr, dims + (T, ), order='F')
Cn2 = cm.local_correlations(Y)
# pl.imshow(Cn2)
#%
#% RUN ALGORITHM ON PATCHES
# pl.close('all')
cnm_init = cnmf.CNMF(n_processes,
method_init='greedy_roi',
k=K,
gSig=gSig,
merge_thresh=merge_thresh,
p=0,
dview=dview,
Ain=None,
rf=rf,
stride=stride,
method_deconvolution='oasis',
skip_refinement=False,
normalize_init=False,
options_local_NMF=None,
minibatch_shape=100,
minibatch_suff_stat=5,
update_num_comps=True,
rval_thr=rval_thr_online,
thresh_fitness_delta=thresh_fitness_delta_online,
thresh_fitness_raw=thresh_fitness_raw_online,
batch_update_suff_stat=True,
max_comp_update_shape=5)
cnm_init = cnm_init.fit(images)
A_tot = cnm_init.A
C_tot = cnm_init.C
YrA_tot = cnm_init.YrA
b_tot = cnm_init.b
f_tot = cnm_init.f
print(('Number of components:' + str(A_tot.shape[-1])))
#%
traces = C_tot + YrA_tot
# traces_a=traces-scipy.ndimage.percentile_filter(traces,8,size=[1,np.shape(traces)[-1]/5])
# traces_b=np.diff(traces,axis=1)
fitness_raw, fitness_delta, erfc_raw, erfc_delta, r_values, significant_samples = evaluate_components(
Y,
traces,
A_tot,
C_tot,
b_tot,
f_tot,
final_frate,
remove_baseline=True,
N=5,
robust_std=False,
Athresh=0.1,
Npeaks=Npeaks,
thresh_C=0.3)
idx_components_r = np.where(r_values >= rval_thr_init)[0]
idx_components_raw = np.where(fitness_raw < thresh_fitness_raw_init)[0]
idx_components_delta = np.where(
fitness_delta < thresh_fitness_delta_init)[0]
idx_components = np.union1d(idx_components_r, idx_components_raw)
idx_components = np.union1d(idx_components, idx_components_delta)
idx_components_bad = np.setdiff1d(list(range(len(traces))), idx_components)
print(('Keeping ' + str(len(idx_components)) + ' and discarding ' +
str(len(idx_components_bad))))
A_tot = A_tot.tocsc()[:, idx_components]
C_tot = C_tot[idx_components]
#%
cnm_refine = cnmf.CNMF(n_processes,
method_init='greedy_roi',
k=A_tot.shape,
gSig=gSig,
merge_thresh=merge_thresh,
rf=None,
stride=None,
p=p,
dview=dview,
Ain=A_tot,
Cin=C_tot,
f_in=f_tot,
method_deconvolution='oasis',
skip_refinement=True,
normalize_init=False,
options_local_NMF=None,
minibatch_shape=100,
minibatch_suff_stat=5,
update_num_comps=True,
rval_thr=rval_thr_refine,
thresh_fitness_delta=thresh_fitness_delta_refine,
thresh_fitness_raw=thresh_fitness_raw_refine,
batch_update_suff_stat=True,
max_comp_update_shape=5)
cnm_refine = cnm_refine.fit(images)
#%
A, C, b, f, YrA, sn = cnm_refine.A, cnm_refine.C, cnm_refine.b, cnm_refine.f, cnm_refine.YrA, cnm_refine.sn
#%
final_frate = 10
Npeaks = 10
traces = C + YrA
fitness_raw, fitness_delta, erfc_raw, erfc_delta, r_values, significant_samples = \
evaluate_components(Y, traces, A, C, b, f, final_frate, remove_baseline=True,
N=5, robust_std=False, Athresh=0.1, Npeaks=Npeaks, thresh_C=0.3)
idx_components_r = np.where(r_values >= rval_thr_refine)[0]
idx_components_raw = np.where(fitness_raw < thresh_fitness_raw_refine)[0]
idx_components_delta = np.where(
fitness_delta < thresh_fitness_delta_refine)[0]
idx_components = np.union1d(idx_components_r, idx_components_raw)
idx_components = np.union1d(idx_components, idx_components_delta)
idx_components_bad = np.setdiff1d(list(range(len(traces))), idx_components)
print(' ***** ')
print((len(traces)))
print((len(idx_components)))
#%
cnm_refine.idx_components = idx_components
cnm_refine.idx_components_bad = idx_components_bad
cnm_refine.r_values = r_values
cnm_refine.fitness_raw = fitness_raw
cnm_refine.fitness_delta = fitness_delta
cnm_refine.Cn2 = Cn2
#%
# cnm_init.dview = None
# save_object(cnm_init,fls[0][:-4]+ '_DS_' + str(ds)+ '_init.pkl')
return cnm_refine, Cn2, fname_new
def run(work_dir: str, UUID: str, save_temp_files: str):
logging.basicConfig(
stream=sys.stdout,
level=logging.DEBUG,
format=
"%(relativeCreated)12d [%(filename)s:%(funcName)20s():%(lineno)s] [%(process)d] %(message)s"
)
start_time = time()
batch_dir = os.environ['CURR_BATCH_DIR']
save_temp_files = bool(int(save_temp_files))
output = {'status': 0, 'output_info': ''}
n_processes = int(os.environ['_MESMERIZE_N_THREADS'])
filepath = os.path.join(work_dir, UUID)
imgpath = f'{filepath}_input.tiff'
input_params = pickle.load(open(f'{filepath}.params', 'rb'))
print('******** Creating process pool *********')
c, dview, n_processes = setup_cluster(backend='local',
n_processes=n_processes,
single_thread=False,
ignore_preexisting=True)
try:
if input_params['use_memmap']:
memmap_uuid = input_params['memmap_uuid']
memmap_batchdir = glob(
os.path.join(batch_dir, f'memmap-{memmap_uuid}*.mmap'))
# Check batch dir
if len(memmap_batchdir) > 0:
memmap_path = memmap_batchdir[0]
print(
f'********** Found existing memmap in batch dir: {memmap_path} ********** '
)
# copy to work dir
if not os.path.samefile(batch_dir, work_dir):
print('**** Copying memmap to work dir ****')
shutil.copy(memmap_path, work_dir)
memmap_path = glob(
os.path.join(work_dir,
f'memmap-{memmap_uuid}*.mmap'))[0]
else:
# remake the memmap with the same UUID so that future batch items can rely on it
print(
'********** Memmap not found, re-making memmap with the same UUID **********'
)
memmap_path = cm.save_memmap([imgpath],
base_name=f'memmap-{memmap_uuid}',
is_3D=True,
order='C',
dview=dview)
else:
print('********** Making memmap **********')
memmap_path = cm.save_memmap([imgpath],
base_name=f'memmap-{UUID}',
is_3D=True,
order='C',
dview=dview)
print(f'Using memmap:\n{memmap_path}')
print('********** Loading memmap **********')
Yr, dims, T = cm.load_memmap(memmap_path)
Y = np.reshape(Yr, dims + (T, ), order='F')
images = np.reshape(Yr.T, [T] + list(dims), order='F')
if input_params['use_patches']:
cnm = cnmf.CNMF(n_processes=n_processes,
dview=dview,
only_init_patch=True,
**input_params['cnmf_kwargs'])
else:
cnm = cnmf.CNMF(n_processes,
dview=dview,
**input_params['cnmf_kwargs'])
cnm.fit(images)
print('Number of components:' + str(cnm.estimates.A.shape[-1]))
cnm.params.change_params(
params_dict={
**input_params['eval_kwargs'], 'use_cnn': False
})
cnm.estimates.evaluate_components(images, cnm.params, dview=dview)
if input_params['refit']:
cnm.params.set('temporal', {'p': input_params['cnmf_kwargs']['p']})
cnm_ = cnm.refit(images)
else:
cnm_ = cnm
out_filename = f'{UUID}_results.hdf5'
cnm_.save(out_filename)
output_files = [out_filename]
# Save the memmap
if save_temp_files:
print("***** Keeping memmap file *****")
# copy to batch dir if batch_dir != work_dir
if not os.path.samefile(batch_dir, work_dir):
print("***** Copying memmap file to batch dir *****")
shutil.copy(memmap_path, batch_dir)
# Delete the memmap from the work dir
if not os.path.samefile(batch_dir, work_dir):
print("***** Deleting memmap files from work dir *****")
try:
os.remove(memmap_path)
except:
pass
output.update({
'output': UUID,
'status': 1,
'output_files': output_files,
'saved_memmap': save_temp_files
})
except Exception as e:
output.update({'status': 0, 'output_info': traceback.format_exc()})
cm.stop_server(dview=dview)
end_time = time()
processing_time = (end_time - start_time) / 60
output.update({'processing_time': processing_time})
json.dump(output, open(filepath + '.out', 'w'))
def run(
file_path,
n_cpus,
motion_correct: bool = True,
quality_control: bool = False,
mc_settings: dict = {},
cnmf_settings: dict = {},
qc_settings: dict = {},
job_name: str = "job",
output_directory: str = "",
):
mkl = os.environ.get("MKL_NUM_THREADS")
blas = os.environ.get("OPENBLAS_NUM_THREADS")
vec = os.environ.get("VECLIB_MAXIMUM_THREADS")
print(f"MKL: {mkl}")
print(f"blas: {blas}")
print(f"vec: {vec}")
# we import the pipeline upon running so they aren't required for all installs
import caiman as cm
from caiman.source_extraction.cnmf import params as params
from caiman.source_extraction import cnmf
# print the directory caiman is imported from
caiman_path = os.path.abspath(cm.__file__)
print(f"caiman path: {caiman_path}")
sys.stdout.flush()
# setup the logger
logger_file = os.path.join(output_directory, "caiman.log")
logging.basicConfig(
format=LOGGER_FORMAT,
filename=logger_file,
filemode="w",
level=logging.DEBUG,
)
# if indices to perform mcorr are set, format them
if "indices" in mc_settings:
indices = mc_settings["indices"]
indices_formatted = ()
for axis_slice in indices:
start = axis_slice[0]
stop = axis_slice[1]
if len(axis_slice) == 3:
step = axis_slice[2]
else:
step = 1
indices_formatted += (slice(start, stop, step), )
mc_settings["indices"] = indices_formatted
# load and update the pipeline settings
mc_parameters = DEFAULT_MCORR_SETTINGS
for k, v in mc_settings.items():
mc_parameters[k] = v
cnmf_parameters = DEFAULT_CNMF_PARAMETERS
for k, v in cnmf_settings.items():
cnmf_parameters[k] = v
qc_parameters = DEFAULT_QC_PARAMETERS
for k, v in qc_settings.items():
qc_parameters[k] = v
opts = params.CNMFParams(params_dict=mc_parameters)
opts.change_params(params_dict=cnmf_parameters)
opts.change_params(params_dict=qc_parameters)
# get the filenames
if os.path.isfile(file_path):
print(file_path)
fnames = [file_path]
else:
file_pattern = os.path.join(file_path, "*.tif*")
fnames = sorted(glob.glob(file_pattern))
print(fnames)
opts.set("data", {"fnames": fnames})
if n_cpus > 1:
print("starting server")
# start the server
n_proc = np.max([(n_cpus - 1), 1])
c, dview, n_processes = cm.cluster.setup_cluster(backend="local",
n_processes=n_proc,
single_thread=False)
print(n_processes)
sleep(30)
else:
print("multiprocessing disabled")
dview = None
n_processes = 1
print("starting analysis")
print(f"perform motion correction: {motion_correct}")
print(f"perform qc: {quality_control}")
sys.stdout.flush()
cnm = cnmf.CNMF(n_processes, params=opts, dview=dview)
cnm_results = cnm.fit_file(motion_correct=motion_correct,
include_eval=quality_control)
print("evaluate components")
sys.stdout.flush()
Yr, dims, T = cm.load_memmap(cnm_results.mmap_file)
images = Yr.T.reshape((T, ) + dims, order="F")
cnm_results.estimates.evaluate_components(images, cnm.params, dview=dview)
print("Number of total components: ", len(cnm_results.estimates.C))
print("Number of accepted components: ",
len(cnm_results.estimates.idx_components))
# save the results object
print("saving results")
results_filebase = os.path.join(output_directory, job_name)
cnm_results.save(results_filebase + ".hdf5")
# if motion correction was performed, save the file
# we save as hdf5 for better reading performance
# downstream
if motion_correct:
print("saving motion corrected file")
mcorr_fname = results_filebase + "_mcorr.hdf5"
dataset_name = cnm_results.params.data["var_name_hdf5"]
fnames = cnm_results.params.data["fnames"]
memmap_files = []
for f in fnames:
if isinstance(f, bytes):
f = f.decode()
base_file = os.path.splitext(f)[0]
if cnm_results.params.motion["pw_rigid"]:
memmap_pattern = base_file + "*_els_*"
else:
memmap_pattern = base_file + "*_rig_*"
memmap_files += sorted(glob.glob(memmap_pattern))
write_hdf5_movie(
movie_name=mcorr_fname,
memmap_files=memmap_files,
frame_shape=cnm_results.dims,
dataset_name=dataset_name,
compression="gzip",
)
# save the parameters in the same dir as the results
final_params = cnm.params.to_dict()
params_file = os.path.join(output_directory, "all_caiman_parameters.pkl")
with open(params_file, "wb") as fp:
pickle.dump(final_params, fp)
print("stopping server")
cm.stop_server(dview=dview)
cnm = cnmf.CNMF(
n_processes=n_processes,
method_init='corr_pnr', # use this for 1 photon
k=70, # neurons per patch
gSig=(3, 3), # half size of neuron
gSiz=(10, 10), # in general 3*gSig+1
merge_thresh=.8, # threshold for merging
p=1, # order of autoregressive process to fit
dview=dview, # if None it will run on a single thread
tsub=
2, # downsampling factor in time for initialization, increase if you have memory problems
ssub=
2, # downsampling factor in space for initialization, increase if you have memory problems
Ain=
None, # if you want to initialize with some preselcted components you can pass them here as boolean vectors
rf=(40, 40), # half size of the patch (final patch will be 100x100)
stride=(
20, 20
), # overlap among patches (keep it at least large as 4 times the neuron size)
only_init_patch=True, # just leave it as is
gnb=16, # number of background components
nb_patch=16, # number of background components per patch
method_deconvolution='oasis', #could use 'cvxpy' alternatively
low_rank_background=True, #leave as is
update_background_components=
True, # sometimes setting to False improve the results
min_corr=min_corr, # min peak value from correlation image
min_pnr=min_pnr, # min peak to noise ration from PNR image
normalize_init=False, # just leave as is
center_psf=True, # leave as is for 1 photon
del_duplicates=True) # whether to remove duplicates from initialization
def process_data(haussio_data,
mask=None,
p=2,
nrois_init=400,
roi_iceberg=0.9,
merge_unconnected=None):
if mask is not None:
raise RuntimeError("mask not supported in cnmf.process_data")
fn_cnmf = haussio_data.dirname_comp + '_cnmf.mat'
shapefn = os.path.join(haussio_data.dirname_comp,
haussio.THOR_RAW_FN[:-3] + "shape.npy")
shape = np.load(shapefn)
if len(shape) == 5:
d1, d2 = shape[2], shape[3]
else:
d1, d2 = shape[1], shape[2]
fn_mmap = get_mmap_name(haussio_data.dirname_comp + os.path.sep + 'Yr', d1,
d2, shape[0])
tiffs_to_cnmf(haussio_data)
if os.path.exists(fn_cnmf):
resdict = loadmat(fn_cnmf)
if "dFoF" in resdict.keys():
A2 = resdict["A"]
C2 = resdict["C"]
YrA = resdict["YrA"]
S2 = resdict["S"]
dFoF = resdict["dFoF"]
bl2 = resdict["bl"]
f = resdict["f"]
images = haussio_data.read_raw().squeeze()
else:
dFoF = None
if not os.path.exists(fn_cnmf) or dFoF is None:
c, dview, n_processes = cm.cluster.setup_cluster(
backend='multiprocessing', n_processes=NCPUS, single_thread=False)
Yr, dims, T = cm.load_memmap(fn_mmap, 'r+')
d1, d2 = dims
images = np.reshape(Yr.T, [T] + list(dims), order='F')
fr = 1.0 / haussio_data.dt # imaging rate in frames per second\n",
decay_time = 0.4 # length of a typical transient in seconds\n",
# parameters for source extraction and deconvolution\n",
bord_px_els = 32 # maximum shift to be used for trimming against NaNs
p = 1 # order of the autoregressive system\n",
gnb = 2 # number of global background components\n",
merge_thresh = 0.8 # merging threshold, max correlation allowed\n",
rf = int(
np.round(np.sqrt(d1 * d2) / nrois_init)
) # half-size of the patches in pixels. e.g., if rf=25, patches are 50x50\n",
if rf < 16:
rf = 16
stride_cnmf = 6 # amount of overlap between the patches in pixels\n",
npatches = np.round(d1 / (rf * 2) * d2 / (rf * 2))
K = nrois_init / npatches # number of components per patch\n",
if K < 2:
K = 2
print(rf, npatches, K)
gSig = [8, 8] # expected half size of neurons\n",
init_method = 'greedy_roi' # initialization method (if analyzing dendritic data using 'sparse_nmf')\n",
is_dendrites = False # flag for analyzing dendritic data\n",
alpha_snmf = None # sparsity penalty for dendritic data analysis through sparse NMF\n",
# parameters for component evaluation\n",
min_SNR = 2.5 # signal to noise ratio for accepting a component\n",
rval_thr = 0.8 # space correlation threshold for accepting a component\n",
cnn_thr = 0.8 # threshold for CNN based classifier"
cnm = caiman_cnmf.CNMF(n_processes=1,
k=K,
gSig=gSig,
merge_thresh=merge_thresh,
p=0,
dview=dview,
rf=rf,
stride=stride_cnmf,
memory_fact=1,
method_init=init_method,
alpha_snmf=alpha_snmf,
only_init_patch=False,
gnb=gnb,
border_pix=bord_px_els)
cnm = cnm.fit(images)
idx_components, idx_components_bad, SNR_comp, r_values, cnn_preds = \
estimate_components_quality_auto(images, cnm.A, cnm.C, cnm.b, cnm.f,
cnm.YrA, fr, decay_time, gSig, dims,
dview = dview, min_SNR=min_SNR,
r_values_min = rval_thr, use_cnn = False,
thresh_cnn_lowest = cnn_thr)
A_in, C_in, b_in, f_in = cnm.A[:, idx_components], cnm.C[
idx_components], cnm.b, cnm.f
cnm2 = caiman_cnmf.CNMF(n_processes=1,
k=A_in.shape[-1],
gSig=gSig,
p=p,
dview=dview,
merge_thresh=merge_thresh,
Ain=A_in,
Cin=C_in,
b_in=b_in,
f_in=f_in,
rf=None,
stride=None,
gnb=gnb,
method_deconvolution='oasis',
check_nan=True)
cnm2 = cnm2.fit(images)
if merge_unconnected is not None:
idx_merge = []
for nroi, ca_roi in enumerate(cnm2.C):
for nroi_compare_counter, ca_roi_compare in enumerate(
cnm2.C[nroi + 1:]):
nroi_compare = nroi_compare_counter + nroi + 1
if nroi_compare not in idx_merge:
correls = np.correlate(ca_roi,
ca_roi_compare,
mode='same')
correls /= np.sqrt(
np.dot(ca_roi, ca_roi) *
np.dot(ca_roi_compare, ca_roi_compare))
if correls.max() > merge_unconnected:
print("Merging ", nroi_compare)
idx_merge.append(nroi_compare)
idx_no_merge = [
idx for idx in range(cnm2.C.shape[0]) if idx not in idx_merge
]
else:
idx_no_merge = range(cnm2.C.shape[0])
A2 = cnm2.A[:, idx_no_merge].tocsc()
C2 = cnm2.C[idx_no_merge]
YrA = cnm2.YrA[idx_no_merge]
S2 = cnm2.S[idx_no_merge]
dFoF = cnm2.detrend_df_f(frames_window=300)[idx_no_merge]
# A: spatial components (ROIs)
# C: denoised [Ca2+]
# YrA: residuals ("noise", i.e. traces = C+YrA)
# S: Spikes
# f: temporal background
savemat(
fn_cnmf, {
"A": A2,
"C": C2,
"YrA": YrA,
"S": S2,
"dFoF": dFoF,
"bl": cnm2.b,
"f": cnm2.f
})
dview.terminate()
cm.stop_server()
proj_fn = haussio_data.dirname_comp + "_proj.npy"
if not os.path.exists(proj_fn):
zproj = utils.zproject(images)
np.save(proj_fn, zproj)
else:
zproj = np.load(proj_fn)
logfiles = glob.glob("*LOG*")
for logfile in logfiles:
try:
os.unlink(logfile)
except OSError:
pass
polygons = contour(A2, images.shape[1], images.shape[2], thr=roi_iceberg)
rois = ROIList([sima.ROI.ROI(polygons=poly) for poly in polygons])
return rois, C2, zproj, S2, images, YrA
def run():
date_recorded = '200306'
mouse_id = 'M504408'
resolution = (512, 512)
channel = 'green'
data_folder_n = '110_LSNDGCUC_reorged'
imaging_mode = 'deepscope' # '2p' or 'deepscope'
n_process = 6
# ========================= caiman parameters for boutons ================================================
# ============ sutter scope, zoom 4, 5 frames online average, 5 frames offline average ===================
# fr = 2. # frame rate (Hz)
# decay_time = 0.5 # approximate length of transient event in seconds
gSig = (
5, 5
) # expected half size of neurons, (8, 8) for soma at zoom 2 on sutter scope
p = 2 # order of AR indicator dynamics
# min_SNR = 3 # minimum SNR for accepting new components
# rval_thr = 0.80 # correlation threshold for new component inclusion
# ds_factor = 1 # spatial downsampling factor (increases speed but may lose some fine structure)
# gnb = 2 # number of background components
# gSig = tuple(np.ceil(np.array(gSig) / ds_factor).astype('int')) # recompute gSig if downsampling is involved
mot_corr = False # flag for online motion correction
pw_rigid = False # flag for pw-rigid motion correction (slower but potentially more accurate)
# max_shifts_online = np.ceil(10. / ds_factor).astype('int') # maximum allowed shift during motion correction
# sniper_mode = True # flag using a CNN to detect new neurons (o/w space correlation is used)
# init_batch = 200 # number of frames for initialization (presumably from the first file)
expected_comps = 500 # maximum number of expected components used for memory pre-allocation (exaggerate here)
# dist_shape_update = True # flag for updating shapes in a distributed way
# min_num_trial = 10 # number of candidate components per frame
K = 10 # initial number of components
# epochs = 2 # number of passes over the data
show_movie = False # show the movie with the results as the data gets processed
method_init = 'sparse_nmf'
do_merge = False
ssub = 1
tsub = 1
alpha_snmf = 10e1
rolling_sum = False
rf = 256
p_ssub = 1
p_tsub = 1
Ain = None
# method_deconvolution = 'oasis'
border_pix = 0
# ========================= caiman parameters for boutons ================================================
curr_folder = os.path.dirname(os.path.realpath(__file__))
# windows
source_folder = r"\\allen\programs\braintv\workgroups\nc-ophys\Jun\raw_data"
# # ubuntu
# source_folder = "/media/nc-ophys/Jun/raw_data"
data_folder = os.path.join(
source_folder, f'{date_recorded}-{mouse_id}-'
f'{imaging_mode}/{data_folder_n}')
plane_ns = [
f for f in os.listdir(data_folder)
if os.path.isdir(os.path.join(data_folder, f)) and f[:5] == 'plane'
]
plane_ns.sort()
print('planes:')
print('\n'.join(plane_ns))
c, dview, _ = cm.cluster.setup_cluster(backend='local',
n_processes=n_process,
single_thread=False)
for plane_n in plane_ns:
print('\nsegmenting plane: {}'.format(plane_n))
plane_folder = os.path.join(data_folder, plane_n, channel, 'corrected')
os.chdir(plane_folder)
fn = [
f for f in os.listdir(plane_folder)
if len(f) > 16 and f[-5:] == '.mmap'
]
if len(fn) > 1:
print('\n'.join(fn))
raise LookupError('more than one file found.')
elif len(fn) == 0:
raise LookupError('no file found.')
else:
fn = fn[0]
fp = os.path.join(os.path.realpath(plane_folder), fn)
params_dict = {
'fnames': [fp],
# 'fr': fr,
# 'decay_time': decay_time,
'gSig': gSig,
'p': p,
# 'min_SNR': min_SNR,
# 'rval_thr': rval_thr,
# 'ds_factor': ds_factor,
# 'nb': gnb,
'motion_correct': mot_corr,
# 'init_batch': init_batch,
# 'init_method': 'bare',
# 'normalize': True,
'expected_comps': expected_comps,
# 'sniper_mode': sniper_mode,
# 'dist_shape_update': dist_shape_update,
# 'min_num_trial': min_num_trial,
'K': K,
# 'epochs': epochs,
# 'max_shifts_online': max_shifts_online,
'pw_rigid': pw_rigid,
'show_movie': show_movie,
# testing parameters
'method_init': method_init,
'do_merge': do_merge,
'ssub': ssub,
'tsub': tsub,
'alpha_snmf': alpha_snmf,
'rolling_sum': rolling_sum,
'rf': rf,
'p_ssub': p_ssub,
'p_tsub': p_tsub,
# 'Ain': Ain,
# 'method_deconvolution': method_deconvolution,
'border_pix': border_pix
}
opts = cnmf.params.CNMFParams(params_dict=params_dict)
cnm1 = cnmf.CNMF(n_process, params=opts, dview=dview)
cnm1.fit_file(motion_correct=False)
roi_num = cnm1.estimates.A.shape[1]
print('saving ...')
save_f = h5py.File('caiman_segmentation_results.hdf5', 'a')
save_f.create_dataset('masks',
data=np.array(
cnm1.estimates.A.todense()).T.reshape(
(roi_num, resolution[0], resolution[1]),
order='F'),
compression='lzf')
save_f.create_dataset('traces', data=cnm1.estimates.C)
save_f.close()
copyfile(
os.path.join(plane_folder, 'caiman_segmentation_results.hdf5'),
os.path.join(curr_folder, plane_n,
'caiman_segmentation_results.hdf5'))
# %% STOP CLUSTER and clean up log files
# cm.stop_server(dview=dview)
log_files = glob.glob('*_LOG_*')
for log_file in log_files:
os.remove(log_file)
#********************************************************************************************************************************
#********************************************************************************************************************************
#%% obtain initial batch file used for initialization
fname_new = Y[:initbatch].save('demo.mmap', order='C') # memory map file (not needed)
Yr, dims, T = cm.load_memmap(fname_new)
images = np.reshape(Yr.T, [T] + list(dims), order='F')
Cn_init = cm.local_correlations(np.reshape(Yr, dims + (T,), order='F'))
#%% RUN (offline) CNMF algorithm on the initial batch
pl.close('all')
cnm_init = cnmf.CNMF(2, k=K, gSig=gSig, merge_thresh=merge_thresh,
p=p, rf=patch_size//2, stride=stride, skip_refinement=False,
normalize_init=False, options_local_NMF=None,
minibatch_shape=100, minibatch_suff_stat=5,
update_num_comps=True, rval_thr=rval_thr,
thresh_fitness_delta=-50, gnb = gnb,
thresh_fitness_raw=thresh_fitness_raw,
batch_update_suff_stat=True, max_comp_update_shape=max_comp_update_shape)
cnm_init = cnm_init.fit(images)
print(('Number of components:' + str(cnm_init.A.shape[-1])))
if False:
pl.figure()
crd = plot_contours(cnm_init.A.tocsc(), Cn_init, thr=0.9)
#%% run (online) OnACID algorithm
Y = np.reshape(Yr, dims + (T, ), order='F')
Cn2 = cm.local_correlations(Y)
pl.imshow(Cn2)
#%% RUN ALGORITHM ON PATCHES
pl.close('all')
cnm_init = cnmf.CNMF(n_processes,
method_init='greedy_roi',
k=K,
gSig=gSig,
merge_thresh=merge_thresh,
p=0,
dview=dview,
Ain=None,
rf=rf,
stride=stride,
method_deconvolution='oasis',
skip_refinement=False,
normalize_init=False,
options_local_NMF=None,
minibatch_shape=100,
minibatch_suff_stat=5,
update_num_comps=True,
rval_thr=rval_thr,
thresh_fitness_delta=thresh_fitness_delta,
thresh_fitness_raw=thresh_fitness_raw,
batch_update_suff_stat=True,
max_comp_update_shape=5)
cnm_init = cnm_init.fit(images)
A_tot = cnm_init.A
C_tot = cnm_init.C
YrA_tot = cnm_init.YrA
params_dict = {
'fnames': fname_green,
'fr': fr,
'decay_time': decay_time,
'gSig': gSig,
'p': p,
'min_SNR': min_SNR,
'rval_thr': rval_thr,
'nb': gnb,
'only_init': False,
'rf': None
}
opts = cnmf.params.CNMFParams(params_dict=params_dict)
cnm_b = cnmf.CNMF(n_processes, params=opts, dview=dview, Ain=Ain)
cnm_b.fit_file(motion_correct=False)
# %% load memory mapped file and evaluate components
Yr, dims, T = cm.load_memmap(cnm_b.mmap_file)
images = np.reshape(Yr.T, [T] + list(dims), order='F')
Cn = cm.local_correlations(images, swap_dim=False)
cnm_b.estimates.plot_contours(img=Cn)
# %% evalute components and do some plotting
cnm_b.estimates.evaluate_components(images, cnm_b.params, dview=dview)
cnm_b.estimates.plot_contours(img=Cn, idx=cnm_b.estimates.idx_components)
cnm_b.estimates.view_components(images,
img=Cn,
idx=cnm_b.estimates.idx_components)
#%%
rf = 10 # half-size of the patches in pixels. rf=25, patches are 50x50
stride = 4 # amounpl.it of overlap between the patches in pixels
K = 4 # number of neurons expected per patch
gSig = [5, 5] # expected half size of neurons
merge_thresh = 0.8 # merging threshold, max correlation allowed
p = 2 # order of the autoregressive system
memory_fact = 1 # unitless number accounting how much memory should be used. You will need to try different values to see which one would work the default is OK for a 16 GB system
save_results = False
#%% RUN ALGORITHM ON PATCHES
cnm = cnmf.CNMF(n_processes,
k=K,
gSig=gSig,
merge_thresh=0.8,
p=0,
dview=dview,
Ain=None,
rf=rf,
stride=stride,
memory_fact=memory_fact,
method_init='greedy_roi',
alpha_snmf=10e2)
cnm = cnm.fit(images)
A_tot = cnm.A
C_tot = cnm.C
YrA_tot = cnm.YrA
b_tot = cnm.b
f_tot = cnm.f
sn_tot = cnm.sn
print(('Number of components:' + str(A_tot.shape[-1])))
def run(batch_dir: str, UUID: str):
start_time = time()
output = {'status': 0, 'output_info': ''}
n_processes = os.environ['_MESMERIZE_N_THREADS']
n_processes = int(n_processes)
file_path = batch_dir + '/' + UUID
filename = file_path + '.tiff'
input_params = pickle.load(open(file_path + '.params', 'rb'))
frate = input_params['frate']
gSig = input_params['gSig']
gSiz = 3 * gSig + 1
min_corr = input_params['min_corr']
min_pnr = input_params['min_pnr']
min_SNR = input_params['min_SNR']
r_values_min = input_params['r_values_min']
decay_time = input_params['decay_time']
rf = input_params['rf']
stride = input_params['stride']
gnb = input_params['gnb']
nb_patch = input_params['nb_patch']
k = input_params['k']
if 'Ain' in input_params.keys():
if input_params['Ain']:
print('>> Ain specified, looking for cnm-A file <<')
item_uuid = input_params['Ain']
parent_batch_dir = os.environ['CURR_BATCH_DIR']
item_out_file = os.path.join(parent_batch_dir, f'{item_uuid}.out')
t0 = time()
timeout = 60
while not os.path.isfile(item_out_file):
print('>>> cnm-A not found, waiting for 15 seconds <<<')
sleep(15)
if time() - t0 > timeout:
output.update({'status': 0, 'output_info': 'Timeout exceeding in waiting for Ain input file'})
raise TimeoutError('Timeout exceeding in waiting for Ain input file')
if os.path.isfile(item_out_file):
if json.load(open(item_out_file, 'r'))['status']:
Ain_file = os.path.join(parent_batch_dir, item_uuid + '_cnm-A.pikl')
Ain = pickle.load(open(Ain_file, 'rb'))
print('>>> Found Ain file <<<')
else:
raise FileNotFoundError('>>> Could not find specified Ain file <<<')
else:
Ain = None
else:
Ain = None
if 'method_deconvolution' in input_params.keys():
method_deconvolution = input_params['method_deconvolution']
else:
method_deconvolution = 'oasis'
if 'deconv_flag' in input_params.keys():
deconv_flag = input_params['deconv_flag']
else:
deconv_flag = True
filename = [filename]
print('*********** Creating Process Pool ***********')
c, dview, n_processes = cm.cluster.setup_cluster(backend='local', # use this one
n_processes=n_processes,
single_thread=False)
if 'bord_px' in input_params.keys():
bord_px = input_params['bord_px']
else:
bord_px = 6
try:
print('Creating memmap')
fname_new = cm.save_memmap_each(
filename,
base_name='memmap_' + UUID,
order='C',
border_to_0=bord_px,
dview=dview)
fname_new = cm.save_memmap_join(fname_new, base_name='memmap_' + UUID, dview=dview)
# load memory mappable file
Yr, dims, T = cm.load_memmap(fname_new)
Y = Yr.T.reshape((T,) + dims, order='F')
# compute some summary images (correlation and peak to noise)
# change swap dim if output looks weird, it is a problem with tiffile
cn_filter, pnr = cm.summary_images.correlation_pnr(
Y, gSig=gSig, swap_dim=False)
if not input_params['do_cnmfe'] and input_params['do_corr_pnr']:
pickle.dump(cn_filter, open(UUID + '_cn_filter.pikl', 'wb'), protocol=4)
pickle.dump(pnr, open(UUID + '_pnr.pikl', 'wb'), protocol=4)
output_file_list = [UUID + '_pnr.pikl',
UUID + '_cn_filter.pikl',
UUID + '_dims.pikl',
UUID + '.out'
]
output.update({'output': UUID,
'status': 1,
'output_info': 'inspect correlation & pnr',
'output_files': output_file_list
})
dview.terminate()
for mf in glob(batch_dir + '/memmap_*'):
os.remove(mf)
end_time = time()
processing_time = (end_time - start_time) / 60
output.update({'processing_time': processing_time})
json.dump(output, open(file_path + '.out', 'w'))
return
cnm = cnmf.CNMF(n_processes=n_processes,
method_init='corr_pnr', # use this for 1 photon
k=k, # neurons per patch
gSig=(gSig, gSig), # half size of neuron
gSiz=(gSiz, gSiz), # in general 3*gSig+1
merge_thresh=.3, # threshold for merging
p=1, # order of autoregressive process to fit
dview=dview, # if None it will run on a single thread
# downsampling factor in time for initialization, increase if you have memory problems
tsub=2,
# downsampling factor in space for initialization, increase if you have memory problems
ssub=2,
# if you want to initialize with some preselcted components you can pass them here as boolean vectors
Ain=Ain,
# half size of the patch (final patch will be 100x100)
rf=(rf, rf),
# overlap among patches (keep it at least large as 4 times the neuron size)
stride=(stride, stride),
only_init_patch=True, # just leave it as is
gnb=gnb, # number of background components
nb_patch=nb_patch, # number of background components per patch
method_deconvolution=method_deconvolution, # could use 'cvxpy' alternatively
deconv_flag=deconv_flag,
low_rank_background=True, # leave as is
# sometimes setting to False improve the results
update_background_components=True,
min_corr=min_corr, # min peak value from correlation image
min_pnr=min_pnr, # min peak to noise ration from PNR image
normalize_init=False, # just leave as is
center_psf=True, # leave as is for 1 photon
del_duplicates=True, # whether to remove duplicates from initialization
border_pix=bord_px) # number of pixels to not consider in the borders
cnm.fit(Y)
# DISCARD LOW QUALITY COMPONENTS
idx_components, idx_components_bad, comp_SNR, r_values, pred_CNN = estimate_components_quality_auto(
Y, cnm.A, cnm.C, cnm.b, cnm.f, cnm.YrA, frate,
decay_time, gSig, dims, dview=dview,
min_SNR=min_SNR, r_values_min=r_values_min, use_cnn=False)
# np.save(filename[:-5] + '_curves.npy', cnm.C)
pickle.dump(Yr, open(UUID + '_Yr.pikl', 'wb'), protocol=4)
pickle.dump(cnm.A, open(UUID + '_cnm-A.pikl', 'wb'), protocol=4)
pickle.dump(cnm.b, open(UUID + '_cnm-b.pikl', 'wb'), protocol=4)
pickle.dump(cnm.C, open(UUID + '_cnm-C.pikl', 'wb'), protocol=4)
pickle.dump(cnm.f, open(UUID + '_cnm-f.pikl', 'wb'), protocol=4)
pickle.dump(idx_components, open(UUID + '_idx_components.pikl', 'wb'), protocol=4)
pickle.dump(cnm.YrA, open(UUID + '_cnm-YrA.pikl', 'wb'), protocol=4)
pickle.dump(pnr, open(UUID + '_pnr.pikl', 'wb'), protocol=4)
pickle.dump(cn_filter, open(UUID + '_cn_filter.pikl', 'wb'), protocol=4)
pickle.dump(dims, open(UUID + '_dims.pikl', 'wb'), protocol=4)
output_file_list = [UUID + '_cnm-A.pikl',
UUID + '_Yr.pikl',
UUID + '_cnm-b.pikl',
UUID + '_cnm-C.pikl',
UUID + '_cnm-f.pikl',
UUID + '_idx_components.pikl',
UUID + '_cnm-YrA.pikl',
UUID + '_pnr.pikl',
UUID + '_cn_filter.pikl',
UUID + '_dims.pikl',
UUID + '.out'
]
output.update({'output': filename[:-5],
'status': 1,
'output_files': output_file_list
})
except Exception as e:
output.update({'status': 0, 'output_info': traceback.format_exc()})
dview.terminate()
for mf in glob(batch_dir + '/memmap_*'):
os.remove(mf)
end_time = time()
processing_time = (end_time - start_time) / 60
output.update({'processing_time': processing_time})
json.dump(output, open(file_path + '.out', 'w'))
def run_source_extraction(input_file, dview):
"""
This is the function for source extraction.
Its goal is to take in a .mmap file,
perform source extraction on it using cnmf-e and save the cnmf object as a .pkl file.
"""
sql = "SELECT equalization,source_extraction_session_wise,fr,decay_time,min_corr,min_pnr,p,K,gSig,merge_thr,rf,stride,tsub,ssub,p_tsub,p_ssub,low_rank_background,nb,nb_patch,ssub_B,init_iter,ring_size_factor,method_init,method_deconvolution,update_background_components,center_psf,border_pix,normalize_init,del_duplicates,only_init FROM Analysis WHERE motion_correction_main =? OR alignment_main = ? OR equalization_main =?"
val = [input_file, input_file, input_file]
cursor.execute(sql, val)
result = cursor.fetchall()
para = []
inter = []
for x in result:
inter = x
for y in inter:
para.append(y)
gSiz = 4 * para[8] + 1
parameters = {
'equalization': para[0],
'session_wise': para[1],
'fr': para[2],
'decay_time': para[3],
'min_corr': para[4],
'min_pnr': para[5],
'p': para[6],
'K': para[7],
'gSig': (para[8], para[8]),
'gSiz': (gSiz, gSiz),
'merge_thr': para[9],
'rf': para[10],
'stride': para[11],
'tsub': para[12],
'ssub': para[13],
'p_tsub': para[14],
'p_ssub': para[15],
'low_rank_background': para[16],
'nb': para[17],
'nb_patch': para[18],
'ssub_B': para[19],
'init_iter': para[20],
'ring_size_factor': para[21],
'method_init': para[22],
'method_deconvolution': para[23],
'update_background_components': para[24],
'center_psf': para[25],
'border_pix': para[26],
'normalize_init': para[27],
'del_duplicates': para[28],
'only_init': para[29]
}
# Determine output paths
if parameters['session_wise']:
data_dir = os.environ[
'DATA_DIR_LOCAL'] + 'data/interim/source_extraction/session_wise/'
else:
data_dir = os.environ[
'DATA_DIR_LOCAL'] + 'data/interim/source_extraction/trial_wise/'
sql1 = "SELECT mouse,session,trial,is_rest,decoding_v,cropping_v,motion_correction_v,alignment_v,equalization_v,source_extraction_v,input,home_path,decoding_main FROM Analysis WHERE motion_correction_main =? OR alignment_main = ? OR equalization_main =?"
val1 = [input_file, input_file, input_file]
cursor.execute(sql1, val1)
result = cursor.fetchall()
data = []
inter = []
for x in result:
inter = x
for y in inter:
data.append(y)
# Update the database
if data[9] == 0:
data[9] = 1
file_name = f"mouse_{data[0]}_session_{data[1]}_trial_{data[2]}.{data[3]}.v{data[4]}.{data[5]}.{data[6]}.{data[7]}.{data[8]}.{data[9]}"
output_file_path = data_dir + f'main/{file_name}.hdf5'
sql1 = "UPDATE Analysis SET source_extraction_main=?,source_extraction_v=? WHERE motion_correction_main =? OR alignment_main = ? OR equalization_main =? "
val1 = [output_file_path, data[9], input_file, input_file, input_file]
cursor.execute(sql1, val1)
else:
data[9] += 1
file_name = f"mouse_{data[0]}_session_{data[1]}_trial_{data[2]}.{data[3]}.v{data[4]}.{data[5]}.{data[6]}.{data[7]}.{data[8]}.{data[9]}"
output_file_path = data_dir + f'main/{file_name}.hdf5'
sql2 = "INSERT INTO Analysis (source_extraction_main,source_extraction_v) VALUES (?,?)"
val2 = [output_file_path, data[10]]
cursor.execute(sql2, val2)
database.commit()
database.commit()
# Load memmory mappable input file
if os.path.isfile(input_file):
Yr, dims, T = cm.load_memmap(input_file)
images = Yr.T.reshape((T, ) + dims, order='F')
else:
logging.warning(f' .mmap file does not exist. Cancelling')
# SOURCE EXTRACTION
# Check if the summary images are already there
corr_npy_file_path, pnr_npy_file_path = get_corr_pnr_path(
gSig_abs=parameters['gSig'][0])
if corr_npy_file_path != None and os.path.isfile(corr_npy_file_path):
# Already computed summary images
logging.info(f' Already computed summary images')
cn_filter = np.load(corr_npy_file_path)
pnr = np.load(pnr_npy_file_path)
else:
# Compute summary images
t0 = datetime.datetime.today()
logging.info(f' Computing summary images')
cn_filter, pnr = cm.summary_images.correlation_pnr(
images[::1], gSig=parameters['gSig'][0], swap_dim=False)
dt = int((datetime.datetime.today() - t0).seconds /
60) # timedelta in minutes
logging.info(f' Computed summary images. dt = {dt} min')
# Saving summary images as npy files
gSig = parameters['gSig'][0]
corr_npy_file_path = data_dir + f'/meta/corr/{file_name}_gSig_{gSig}.npy'
pnr_npy_file_path = data_dir + f'/meta/pnr/{file_name}_gSig_{gSig}.npy'
with open(corr_npy_file_path, 'wb') as f:
np.save(f, cn_filter)
with open(pnr_npy_file_path, 'wb') as f:
np.save(f, pnr)
# Calculate min, mean, max value for cn_filter and pnr
corr_min, corr_mean, corr_max = cn_filter.min(), cn_filter.mean(
), cn_filter.max()
pnr_min, pnr_mean, pnr_max = pnr.min(), pnr.mean(), pnr.max()
# If min_corr and min_pnr are specified via a linear equation, calculate
# this value
if type(parameters['min_corr']) == list:
min_corr = parameters['min_corr'][0] * corr_mean + parameters[
'min_corr'][1]
parameters['min_corr'] = min_corr
logging.info(f' Automatically setting min_corr = {min_corr}')
if type(parameters['min_pnr']) == list:
min_pnr = parameters['min_pnr'][0] * pnr_mean + parameters['min_pnr'][1]
parameters['min_pnr'] = min_pnr
logging.info(f' Automatically setting min_pnr = {min_pnr}')
# Set the parameters for caiman
opts = params.CNMFParams(params_dict=parameters)
# SOURCE EXTRACTION
logging.info(f' Performing source extraction')
t0 = datetime.datetime.today()
n_processes = psutil.cpu_count()
logging.info(f' n_processes: {n_processes}')
cnm = cnmf.CNMF(n_processes=n_processes, dview=dview, params=opts)
cnm.fit(images)
cnm.estimates.dims = dims
# Calculate the center of masses
cnm.estimates.center = caiman.base.rois.com(cnm.estimates.A,
images.shape[1],
images.shape[2])
# Save the cnmf object as a hdf5 file
logging.info(f' Saving cnmf object')
cnm.save(output_file_path)
dt = int(
(datetime.datetime.today() - t0).seconds / 60) # timedelta in minutes
logging.info(f' Source extraction finished. dt = {dt} min')
sql1 = "UPDATE Analysis SET duration_summary_images=?,source_extraction_corr=?, source_extraction_pnr=?, source_extraction_corr_min =?, source_extraction_corr_mean=?, source_extraction_corr_max=?, source_extraction_pnr_min=?,source_extraction_pnr_mean=?,source_extraction_pnr_max=?,source_extraction_k=?,source_extraction_duration=?,min_corr=?,min_pnr=? WHERE source_extraction_main= ? AND source_extraction_v=? "
val1 = [
dt, corr_npy_file_path, pnr_npy_file_path, corr_min, corr_mean,
corr_max, pnr_min, pnr_mean, pnr_max,
len(cnm.estimates.C), dt, output_file_path, data[9]
]
cursor.execute(sql1, val1)
return output_file_path, data[9]
single_thread=False)
#%%
cnm = cnmf.CNMF(n_processes=n_processes,
method_init='corr_pnr',
k=35,
gSig=(3, 3),
gSiz=(10, 10),
merge_thresh=.8,
p=1,
dview=dview,
tsub=1,
ssub=1,
Ain=None,
rf=(25, 25),
stride=(25, 25),
only_init_patch=True,
gnb=5,
nb_patch=3,
method_deconvolution='oasis',
low_rank_background=False,
update_background_components=False,
min_corr=min_corr,
min_pnr=min_pnr,
normalize_init=False,
deconvolve_options_init=None,
ring_size_factor=1.5,
center_psf=True)
#%%
# cnm = cnmf.CNMF(n_processes=2, method_init='corr_pnr', k=155, gSig=(3, 3), gSiz=(10, 10), merge_thresh=.8,
def run_source_extraction(row, parameters, dview, session_wise = False):
'''
This is the function for source extraction.
Its goal is to take in a .mmap file,
perform source extraction on it using cnmf-e and save the cnmf object as a .pkl file.
Args:
row: pd.DataFrame object
The row corresponding to the analysis state to be source extracted.
Returns:
row: pd.DataFrame object
The row corresponding to the source extracted analysis state.
'''
step_index = 5
row_local = row.copy()
row_local.loc['source_extraction_parameters'] = str(parameters)
row_local = db.set_version_analysis('source_extraction',row_local,session_wise)
index = row_local.name
# Determine input path
if parameters['session_wise']:
input_mmap_file_path = eval(row_local.loc['alignment_output'])['main']
if parameters['equalization']:
input_mmap_file_path =eval(row_local['equalization_output'])['main']
else:
input_mmap_file_path = eval(row_local.loc['motion_correction_output'])['main']
if not os.path.isfile(input_mmap_file_path):
logging.error('Input file does not exist. Cancelling.')
return row_local
# Determine output paths
file_name = db.create_file_name(step_index, index)
if parameters['session_wise']:
data_dir = os.environ['DATA_DIR'] + 'data/interim/source_extraction/session_wise/'
else:
data_dir = os.environ['DATA_DIR'] + 'data/interim/source_extraction/trial_wise/'
output_file_path = data_dir + f'main/{file_name}.hdf5'
# Create a dictionary with parameters
output = {
'main': output_file_path,
'meta':{
'analysis' : {
'analyst' : os.environ['ANALYST'],
'date' : datetime.datetime.today().strftime("%m-%d-%Y"),
'time' : datetime.datetime.today().strftime("%H:%M:%S"),
},
'duration': {}
}
}
# Load memmory mappable input file
if os.path.isfile(input_mmap_file_path):
Yr, dims, T = cm.load_memmap(input_mmap_file_path)
# logging.debug(f'{index} Loaded movie. dims = {dims}, T = {T}.')
images = Yr.T.reshape((T,) + dims, order='F')
else:
logging.warning(f'{index} .mmap file does not exist. Cancelling')
return row_local
# SOURCE EXTRACTION
# Check if the summary images are already there
corr_npy_file_path, pnr_npy_file_path = fm.get_corr_pnr_path(index, gSig_abs = parameters['gSig'][0])
if corr_npy_file_path != None and os.path.isfile(corr_npy_file_path):
# Already computed summary images
logging.info(f'{index} Already computed summary images')
cn_filter = np.load(corr_npy_file_path)
pnr = np.load(pnr_npy_file_path)
else:
# Compute summary images
t0 = datetime.datetime.today()
logging.info(f'{index} Computing summary images')
cn_filter, pnr = cm.summary_images.correlation_pnr(images[::1], gSig = parameters['gSig'][0], swap_dim=False)
dt = int((datetime.datetime.today() - t0).seconds/60) # timedelta in minutes
output['meta']['duration']['summary_images'] = dt
logging.info(f'{index} Computed summary images. dt = {dt} min')
# Saving summary images as npy files
gSig = parameters['gSig'][0]
corr_npy_file_path = data_dir + f'/meta/corr/{db.create_file_name(3, index)}_gSig_{gSig}.npy'
pnr_npy_file_path = data_dir + f'/meta/pnr/{db.create_file_name(3, index)}_gSig_{gSig}.npy'
with open(corr_npy_file_path, 'wb') as f:
np.save(f, cn_filter)
with open(pnr_npy_file_path, 'wb') as f:
np.save(f, pnr)
# Store the paths in the meta dictionary
output['meta']['corr'] = {'main': corr_npy_file_path, 'meta': {}}
output['meta']['pnr'] = {'main': pnr_npy_file_path, 'meta': {}}
# Calculate min, mean, max value for cn_filter and pnr
corr_min, corr_mean, corr_max = cn_filter.min(), cn_filter.mean(), cn_filter.max()
output['meta']['corr']['meta'] = {'min': corr_min, 'mean': corr_mean, 'max': corr_max}
pnr_min, pnr_mean, pnr_max = pnr.min(), pnr.mean(), pnr.max()
output['meta']['pnr']['meta'] = {'min': pnr_min, 'mean': pnr_mean, 'max': pnr_max}
# If min_corr and min_pnr are specified via a linear equation, calculate
# this value
if type(parameters['min_corr']) == list:
min_corr = parameters['min_corr'][0]*corr_mean + parameters['min_corr'][1]
parameters['min_corr'] = min_corr
logging.info(f'{index} Automatically setting min_corr = {min_corr}')
if type(parameters['min_pnr']) == list:
min_pnr = parameters['min_pnr'][0]*pnr_mean + parameters['min_pnr'][1]
parameters['min_pnr'] = min_pnr
logging.info(f'{index} Automatically setting min_pnr = {min_pnr}')
# Set the parameters for caiman
opts = params.CNMFParams(params_dict = parameters)
# SOURCE EXTRACTION
logging.info(f'{index} Performing source extraction')
t0 = datetime.datetime.today()
n_processes = psutil.cpu_count()
logging.info(f'{index} n_processes: {n_processes}')
cnm = cnmf.CNMF(n_processes = n_processes, dview = dview, params = opts)
cnm.fit(images)
cnm.estimates.dims = dims
# Store the number of neurons
output['meta']['K'] = len(cnm.estimates.C)
# Calculate the center of masses
cnm.estimates.center = caiman.base.rois.com(cnm.estimates.A, images.shape[1], images.shape[2])
# Save the cnmf object as a hdf5 file
logging.info(f'{index} Saving cnmf object')
cnm.save(output_file_path)
dt = int((datetime.datetime.today() - t0).seconds/60) # timedelta in minutes
output['meta']['duration']['source_extraction'] = dt
logging.info(f'{index} Source extraction finished. dt = {dt} min')
# Write necessary variables in row and return
row_local.loc['source_extraction_parameters'] = str(parameters)
row_local.loc['source_extraction_output'] = str(output)
return row_local
def run(batch_dir: str, UUID: str):
logging.basicConfig(
stream=sys.stdout,
level=logging.DEBUG,
format=
"%(relativeCreated)12d [%(filename)s:%(funcName)20s():%(lineno)s] [%(process)d] %(message)s"
)
start_time = time()
output = {'status': 0, 'output_info': ''}
n_processes = os.environ['_MESMERIZE_N_THREADS']
n_processes = int(n_processes)
file_path = os.path.join(batch_dir, UUID)
filename = [file_path + '_input.tiff']
input_params = pickle.load(open(file_path + '.params', 'rb'))
# If Ain is specified
if input_params['do_cnmfe']:
Ain = None
item_uuid = input_params['cnmfe_kwargs'].pop('Ain')
if item_uuid:
print('>> Ain specified, looking for cnm-A file <<')
parent_batch_dir = os.environ['CURR_BATCH_DIR']
item_out_file = os.path.join(parent_batch_dir, f'{item_uuid}.out')
t0 = time()
timeout = 60
while not os.path.isfile(item_out_file):
print('>>> cnm-A not found, waiting for 15 seconds <<<')
sleep(15)
if time() - t0 > timeout:
output.update({
'status':
0,
'output_info':
'Timeout exceeding in waiting for Ain input file'
})
raise TimeoutError(
'Timeout exceeding in waiting for Ain input file')
if os.path.isfile(item_out_file):
if json.load(open(item_out_file, 'r'))['status']:
Ain_path = os.path.join(parent_batch_dir,
item_uuid + '_results.hdf5')
Ain = load_dict_from_hdf5(Ain_path)['estimates']['A']
print('>>> Found Ain file <<<')
else:
raise FileNotFoundError(
'>>> Could not find specified Ain file <<<')
print('*********** Creating Process Pool ***********')
c, dview, n_processes = cm.cluster.setup_cluster(backend='local',
n_processes=n_processes,
single_thread=False
#ignore_preexisting=True
)
try:
print('Creating memmap')
# memmap_path = cm.save_memmap_each(
# filename,
# base_name='memmap-' + UUID,
# order='C',
# border_to_0=input_params['border_pix'],
# dview=dview)
# memmap_path = cm.save_memmap_join(memmap_path, base_name='memmap-' + UUID, dview=dview)
# # load memory mappable file
# Yr, dims, T = cm.load_memmap(memmap_path)
# Y = Yr.T.reshape((T,) + dims, order='F')
memmap_path = cm.save_memmap(
filename,
base_name=f'memmap-',
order='C',
dview=dview,
border_to_0=input_params['border_pix'],
)
Yr, dims, T = cm.load_memmap(memmap_path)
Y = np.reshape(Yr.T, [T] + list(dims), order='F')
if input_params['do_cnmfe']:
gSig = input_params['cnmfe_kwargs']['gSig'][0]
else:
gSig = input_params['corr_pnr_kwargs']['gSig']
cn_filter, pnr = cm.summary_images.correlation_pnr(Y,
swap_dim=False,
gSig=gSig)
if not input_params['do_cnmfe'] and input_params['do_corr_pnr']:
pickle.dump(cn_filter,
open(UUID + '_cn_filter.pikl', 'wb'),
protocol=4)
pickle.dump(pnr, open(UUID + '_pnr.pikl', 'wb'), protocol=4)
output_file_list = \
[
UUID + '_pnr.pikl',
UUID + '_cn_filter.pikl',
]
output.update({
'output': UUID,
'status': 1,
'output_info': 'inspect correlation & pnr',
'output_files': output_file_list
})
dview.terminate()
for mf in glob(batch_dir + '/memmap-*'):
os.remove(mf)
end_time = time()
processing_time = (end_time - start_time) / 60
output.update({'processing_time': processing_time})
json.dump(output, open(file_path + '.out', 'w'))
return
cnm = cnmf.CNMF(
n_processes=n_processes,
method_init='corr_pnr',
dview=dview,
Ain=Ain,
only_init_patch=True, # just leave it as is
normalize_init=False,
center_psf=True,
**input_params['cnmfe_kwargs'],
)
cnm.fit(Y)
# DISCARD LOW QUALITY COMPONENTS
cnm.params.set('quality', {
'use_cnn': False,
**input_params['eval_kwargs']
})
cnm.estimates.evaluate_components(Y, cnm.params, dview=dview)
out_filename = f'{UUID}_results.hdf5'
cnm.save(out_filename)
pickle.dump(pnr, open(UUID + '_pnr.pikl', 'wb'), protocol=4)
pickle.dump(cn_filter,
open(UUID + '_cn_filter.pikl', 'wb'),
protocol=4)
output.update({
'output':
filename[:-5],
'status':
1,
'output_files': [
out_filename,
UUID + '_pnr.pikl',
UUID + '_cn_filter.pikl',
]
})
except Exception as e:
output.update({
'status': 0,
'Y.shape': Y.shape,
'output_info': traceback.format_exc()
})
dview.terminate()
for mf in glob(batch_dir + '/memmap-*'):
try:
os.remove(mf)
except: # Windows doesn't like removing the memmaps
pass
end_time = time()
processing_time = (end_time - start_time) / 60
output.update({'processing_time': processing_time})
json.dump(output, open(file_path + '.out', 'w'))
Yr, dims, T = cm.load_memmap(fname_new)
Y = Yr.T.reshape((T, ) + dims, order='F')
#%% run w/o patches
dview, n_processes = None, 2
cnm = cnmf.CNMF(n_processes=n_processes,
method_init='corr_pnr',
k=None,
dview=dview,
gSig=(gSig, gSig),
gSiz=(gSiz, gSiz),
merge_thresh=.65,
p=1,
tsub=1,
ssub=1,
only_init_patch=True,
gnb=0,
min_corr=.7,
min_pnr=7,
normalize_init=False,
ring_size_factor=1.4,
center_psf=True,
ssub_B=2,
init_iter=1,
s_min=-10)
cnm.fit(Y)
#%% run w/ patches
c, dview, n_processes = cm.cluster.setup_cluster(backend='local',
n_processes=None,
single_thread=False)
K = 14 # number of neurons expected per patch
gSig = [7, 7] # expected half size of neurons
merge_thresh = 0.8 # merging threshold, max correlation allowed
p = 1 # order of the autoregressive system
memory_fact = 1 # unitless number accounting how much memory should be used. You will need to try different values to see which one would work the default is OK for a 16 GB system
save_results = False
#%% RUN ALGORITHM ON PATCHES
t1 = time()
cnm = cnmf.CNMF(n_processes,
k=K,
gSig=gSig,
merge_thresh=0.8,
p=0,
dview=dview,
Ain=None,
rf=rf,
stride=stride,
memory_fact=memory_fact,
method_init=init_method,
alpha_snmf=alpha_snmf,
only_init_patch=True,
gnb=1,
method_deconvolution='oasis')
cnm = cnm.fit(images)
A_tot = cnm.A
C_tot = cnm.C
YrA_tot = cnm.YrA
b_tot = cnm.b
f_tot = cnm.f
sn_tot = cnm.sn
'nb': gnb,
'nb_patch': nb_patch,
'method_deconvolution': 'oasis', # could use 'cvxpy' alternatively
'low_rank_background': low_rank_background,
'update_background_components': True, # sometimes setting to False improve the results
'min_corr': min_corr,
'min_pnr': min_pnr,
'normalize_init': False, # just leave as is
'center_psf': True, # leave as is for 1 photon
'ssub_B': ssub_B,
'ring_size_factor': ring_size_factor,
'del_duplicates': True, # whether to remove duplicates from initialization
'border_pix': 10}) # number of pixels to not consider in the borders)
cnm = cnmf.CNMF(n_processes=n_processes, dview=dview, Ain=Ain, params=opts)
cnm.fit(images)
#%% COMPONENT EVALUATION
# the components are evaluated in two ways:
# a) the shape of each component must be correlated with the data
# b) a minimum peak SNR is required over the length of a transient
# Note that here we do not use the CNN based classifier, because it was trained on 2p not 1p data
min_SNR = 2 # adaptive way to set threshold on the transient size
r_values_min = 0.75 # threshold on space consistency (if you lower more components
# will be accepted, potentially with worst quality)
cnm.params.set('quality', {'min_SNR': min_SNR,
'rval_thr': r_values_min,
'use_cnn': False})
cnm.estimates.evaluate_components(images, cnm.params, dview=dview)
# pick thresholds
inspect_correlation_pnr(cn_filter,pnr)
#%% start cluster
try:
dview.terminate()
dview = None
except:
pass
c, dview, n_processes = cm.cluster.setup_cluster(
backend='local', n_processes=None, single_thread=False)
#%%
cnm = cnmf.CNMF(n_processes=n_processes, method_init='corr_pnr', k=10,
gSig=gSig, gSiz=gSiz, merge_thresh=.8, p=1, dview=dview,
tsub=1, ssub=1, Ain=None, rf=(50, 50), stride=(32, 32), only_init_patch=True,
gnb=16, nb_patch=16, method_deconvolution='oasis', low_rank_background=True,
update_background_components=True, min_corr=min_corr, min_pnr=min_pnr,
normalize_init=False, deconvolve_options_init=None,
ring_size_factor=1.5, center_psf=center_psf, del_duplicates=True)
cnm.fit(Y)
# %% DISCARD LOW QUALITY COMPONENT
final_frate = 10
r_values_min = 0.9 # threshold on space consistency
fitness_min = -100 # threshold on time variability
# threshold on time variability (if nonsparse activity)
fitness_delta_min = - 100
Npeaks = 5
traces = cnm.C + cnm.YrA
# TODO: todocument
idx_components, idx_components_bad = cm.components_evaluation.estimate_components_quality(
images = np.reshape(Yr.T, [T] + list(dims), order='F')
Y = np.reshape(Yr, dims + (T, ), order='F')
#%%
Cn = cnmf.utilities.local_correlations(Y[:, :, :3000])
pl.imshow(Cn, cmap='gray')
#%%
if not is_patches:
#%%
K = 35 # number of neurons expected per patch
gSig = [7, 7] # expected half size of neurons
merge_thresh = 0.8 # merging threshold, max correlation allowed
p = 2 #order of the autoregressive system
cnm=cnmf.CNMF(n_processes, method_init=init_method, k=K,gSig=gSig,merge_thresh=merge_thresh,\
p=p,dview=dview,Ain=None)
cnm = cnm.fit(images)
#%%
else:
#%%
rf = 15 # half-size of the patches in pixels. rf=25, patches are 50x50
stride = 4 #amounpl.it of overlap between the patches in pixels
K = 6 # number of neurons expected per patch
gSig = [7, 7] # expected half size of neurons
merge_thresh = 0.8 # merging threshold, max correlation allowed
p = 2 #order of the autoregressive system
memory_fact = 1
#unitless number accounting how much memory should be used. You will need to try different values to see which one would work the default is OK for a 16 GB system
save_results = False
#%% RUN ALGORITHM ON PATCHES
def main():
pass # For compatibility between running under Spyder and the CLI
#%% load data
fname = os.path.join(caiman_datadir(), 'example_movies', 'demoMovie.tif')
Y = cm.load(fname).astype(np.float32) #
# used as a background image
Cn = cm.local_correlations(Y.transpose(1, 2, 0))
#%% set up some parameters
# frame rate (Hz)
fr = 10
# approximate length of transient event in seconds
decay_time = 0.5
# expected half size of neurons
gSig = [6, 6]
# order of AR indicator dynamics
p = 1
# minimum SNR for accepting new components
min_SNR = 3.5
# correlation threshold for new component inclusion
rval_thr = 0.90
# number of background components
gnb = 3
# set up some additional supporting parameters needed for the algorithm (these are default values but change according to dataset characteristics)
# number of shapes to be updated each time (put this to a finite small value to increase speed)
max_comp_update_shape = np.inf
# maximum number of expected components used for memory pre-allocation (exaggerate here)
expected_comps = 50
# number of timesteps to consider when testing new neuron candidates
N_samples = np.ceil(fr * decay_time)
# exceptionality threshold
thresh_fitness_raw = log_ndtr(-min_SNR) * N_samples
# total length of file
T1 = Y.shape[0]
# set up CNMF initialization parameters
# merging threshold, max correlation allowed
merge_thresh = 0.8
# number of frames for initialization (presumably from the first file)
initbatch = 400
# size of patch
patch_size = 32
# amount of overlap between patches
stride = 3
# max number of components in each patch
K = 4
#%% obtain initial batch file used for initialization
# memory map file (not needed)
fname_new = Y[:initbatch].save(os.path.join(caiman_datadir(),
'example_movies', 'demo.mmap'),
order='C')
Yr, dims, T = cm.load_memmap(fname_new)
images = np.reshape(Yr.T, [T] + list(dims), order='F')
Cn_init = cm.local_correlations(np.reshape(Yr, dims + (T, ), order='F'))
#%% RUN (offline) CNMF algorithm on the initial batch
pl.close('all')
cnm_init = cnmf.CNMF(2,
k=K,
gSig=gSig,
merge_thresh=merge_thresh,
fr=fr,
p=p,
rf=patch_size // 2,
stride=stride,
skip_refinement=False,
normalize_init=False,
options_local_NMF=None,
minibatch_shape=100,
minibatch_suff_stat=5,
update_num_comps=True,
rval_thr=rval_thr,
thresh_fitness_delta=-50,
gnb=gnb,
decay_time=decay_time,
thresh_fitness_raw=thresh_fitness_raw,
batch_update_suff_stat=False,
max_comp_update_shape=max_comp_update_shape,
expected_comps=expected_comps,
dview=None,
min_SNR=min_SNR)
cnm_init = cnm_init.fit(images)
print(('Number of components:' + str(cnm_init.estimates.A.shape[-1])))
pl.figure()
crd = plot_contours(cnm_init.estimates.A.tocsc(), Cn_init, thr=0.9)
#%% run (online) OnACID algorithm
cnm = deepcopy(cnm_init)
cnm.params.data['dims'] = (60, 80)
cnm._prepare_object(np.asarray(Yr), T1)
t = initbatch
Y_ = cm.load(fname)[initbatch:].astype(np.float32)
for frame_count, frame in enumerate(Y_):
cnm.fit_next(t, frame.copy().reshape(-1, order='F'))
t += 1
#%% extract the results
C, f = cnm.estimates.C_on[gnb:cnm.M], cnm.estimates.C_on[:gnb]
A, b = cnm.estimates.Ab[:, gnb:cnm.M], cnm.estimates.Ab[:, :gnb]
print(('Number of components:' + str(A.shape[-1])))
#%% pass through the CNN classifier with a low threshold (keeps clearer neuron shapes and excludes processes)
use_CNN = True
if use_CNN:
# threshold for CNN classifier
thresh_cnn = 0.1
from caiman.components_evaluation import evaluate_components_CNN
predictions, final_crops = evaluate_components_CNN(
A,
dims,
gSig,
model_name=os.path.join(caiman_datadir(), 'model', 'cnn_model'))
A_exclude, C_exclude = A[:, predictions[:, 1] < thresh_cnn], C[
predictions[:, 1] < thresh_cnn]
A, C = A[:,
predictions[:,
1] >= thresh_cnn], C[predictions[:,
1] >= thresh_cnn]
noisyC = cnm.estimates.noisyC[gnb:cnm.M]
YrA = noisyC[predictions[:, 1] >= thresh_cnn] - C
else:
YrA = cnm.estimates.noisyC[gnb:cnm.M] - C
#%% plot results
pl.figure()
crd = cm.utils.visualization.plot_contours(A, Cn, thr=0.9)
view_patches_bar(Yr, A, C, b, f, dims[0], dims[1], YrA, img=Cn)
def run_cnmfe(
file_path: str,
n_cpus: int = 1,
mc_settings: dict = {},
cnmf_settings: dict = {},
qc_settings: dict = {},
):
caiman_path = os.path.abspath(cm.__file__)
print(f'caiman location: {caiman_path}')
sys.stdout.flush()
# load and update the pipeline settings
mc_parameters = DEFAULT_MCORR_SETTINGS
for k, v in mc_settings.items():
mc_parameters[k] = v
cnmf_parameters = DEFAULT_CNMF_PARAMETERS
for k, v in cnmf_settings.items():
cnmf_parameters[k] = v
qc_parameters = DEFAULT_QC_PARAMETERS
for k, v in qc_settings.items():
qc_parameters[k] = v
opts = params.CNMFParams(params_dict=mc_parameters)
opts.change_params(params_dict=cnmf_parameters)
opts.change_params(params_dict=qc_parameters)
# add the image files to the data param
opts.set('data', {'fnames': [file_path]})
if n_cpus > 1:
print("starting server")
n_proc = np.max([(n_cpus - 1), 1])
c, dview, n_processes = cm.cluster.setup_cluster(backend="local",
n_processes=n_proc,
single_thread=False)
print(n_processes)
sleep(30)
else:
print('no multiprocessing')
dview = None
n_processes = 1
print('starting cnmfe')
sys.stdout.flush()
cnm = cnmf.CNMF(n_processes, params=opts, dview=dview)
cnm.fit_file(motion_correct=False)
print("saving results")
cnm.save(cnm.mmap_file[:-4] + "hdf5")
# save the parameters in the same dir as the results
final_params = cnm.params.to_dict()
path_base = os.path.dirname(cnm.mmap_file)
params_file = os.path.join(path_base, "all_caiman_parameters.pkl")
with open(params_file, "wb") as fp:
pickle.dump(final_params, fp)
print("stopping server")
cm.stop_server(dview=dview)
