-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmain.py
180 lines (148 loc) · 5.8 KB
/
main.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
# main.py
"""
Created on Thu Sep 9 14:55:12 2021
@author: Mels
"""
import matplotlib.pyplot as plt
import copy
import time
import numpy as np
import tensorflow as tf
tf.get_logger().setLevel('ERROR')
from dataset import dataset
from dataset_simulation import dataset_simulation, dataset_copy, Deform, Affine_Deform
plt.close('all')
DS2=None
#%% Load datasets
if False: #% Load Beads
maxDistance=None
DS1 = dataset(['C:/Users/Mels/Documents/example_MEP/mol115_combined_clusters.hdf5'],
pix_size=159, loc_error=1.4, mu=0,
linked=True, FrameLinking=False, BatchOptimization=False, execute_linked=True)
DS1.load_dataset_hdf5(align_rcc=False)
## optimization params
learning_rate = 1e-2
epochs = 300
pair_filter = [None, 30, 30]
gridsize=4000
if True: #% Load Excel Niekamp
opt_fn=tf.optimizers.SGD
DS1 = dataset('C:/Users/Mels/Documents/Supplementary-data/data/Registration/Set1/set1_beads_locs.csv',
pix_size=1, loc_error=1.4, mu=0.3, linked=False,
FrameLinking=True, BatchOptimization=False, execute_linked=True)
DS2 = dataset('C:/Users/Mels/Documents/Supplementary-data/data/Registration/Set2/set2_beads_locs.csv',
pix_size=1, loc_error=1.4, mu=0.3, linked=False,
FrameLinking=True, execute_linked=True)
DS1.load_dataset_excel()
DS2.load_dataset_excel()
DS1.pix_size=159
DS2.pix_size=DS1.pix_size
## optimization params
learning_rate=1e-3
epochs = 300
pair_filter = [250, 30, 15]
gridsize=6500
# linking and aligning via AffineLLS
maxDistance=1000
k=1
DS1.link_dataset(maxDistance=maxDistance)
DS2.link_dataset(maxDistance=maxDistance)
DS1.AffineLLS(maxDistance, k)
DS2.Apply_Affine(DS1.AffineMat)
DS1.Filter(pair_filter[0])
if True: #% Load FRET clusters
maxDistance=300
k=8
opt_fn=tf.optimizers.SGD
DS1 = dataset(['C:/Users/Mels/Documents/example_MEP/ch0_locs_picked_clusters.hdf5',
'C:/Users/Mels/Documents/example_MEP/ch1_locs_picked_clusters.hdf5'],
pix_size=159, loc_error=10, mu=0, imgshape=[256,512],
linked=False, FrameLinking=True, BatchOptimization=False, execute_linked=False)
DS1.load_dataset_hdf5(align_rcc=True, transpose=False)
DS1, DS2=DS1.SplitDatasetClusters()
DS1clust=DS1.ClusterDataset(loc_error=None)
DS1clust.execute_linked=True
DS2clust=DS2.ClusterDataset(loc_error=None)
DS1clust.link_dataset(maxDistance=maxDistance)
## optimization params
learning_rate=5e-5
epochs=1000
pair_filter=[None, None, maxDistance]
gridsize=7500
#% aligning clusters
DS1clust.AffineLLS(maxDistance, k)
DS1.copy_models(DS1clust) ## Copy all mapping parameters
DS1.Apply_Affine(DS1clust.AffineMat)
if DS2clust is not None:
DS2.copy_models(DS1clust) ## Copy all mapping parameters
DS2.Apply_Affine(DS1clust.AffineMat)
if DS2.SplinesModel is not None: DS2.Apply_Splines()
#% linking dataset
#if not DS1.Neighbours: DS1.kNearestNeighbour(k=k, maxDistance=maxDistance)
#DS1.AffineLLS(maxDistance, k)
#DS2.Apply_Affine(DS1.AffineMat)
DS1.Filter(pair_filter[0])
if False: #% Load DNA-paint
maxDistance=300
k=1
DS1locs = dataset(['C:/Users/Mels/Documents/DNA_PAINT/DNA_PAINT-chan1_picked.hdf5',
'C:/Users/Mels/Documents/DNA_PAINT/DNA_PAINT-chan2_picked.hdf5'],
pix_size=159, loc_error=10, mu=0, imgshape=[512,256],
linked=False, FrameLinking=False, BatchOptimization=False, execute_linked=True)
DS1locs.load_dataset_hdf5(align_rcc=True, transpose=False)
#DS1locs, DS2locs=DS1locs.SplitDatasetClusters()
DS1=DS1locs.ClusterDataset(loc_error=None)
#DS2=DS2locs.ClusterDataset()
DS1.link_dataset(maxDistance=maxDistance)
DS1,DS2=DS1.SplitDataset()
## optimization params
learning_rate=5e-5
epochs=None
pair_filter=[None, None, maxDistance]
gridsize=1000
fig,ax=DS1.show_channel(DS1.ch1.pos, ps=7)
DS1.show_channel(DS1.ch2.pos, ps=7, color='blue',fig=fig, ax=ax)
#%% running the CatmullRomSplines
start=time.time()
if epochs is not None:
DS1.execute_linked=True
if not DS1.linked: DS1.link_dataset(maxDistance=maxDistance)
DS1.Train_Splines(learning_rate, 300, gridsize, edge_grids=1, opt_fn=opt_fn,
maxDistance=maxDistance, k=k)
DS1.Apply_Splines()
DS1.Filter(pair_filter[1])
print('Optimized in',round((time.time()-start)/60,1),'minutes!')
if DS2 is not None:
DS2.copy_models(DS1) ## Copy all mapping parameters
if DS2.SplinesModel is not None: DS2.Apply_Splines()
#%% output
nbins=100
xlim=pair_filter[2]
if not DS1.linked:
DS1.link_dataset(maxDistance=maxDistance)
## DS1
#DS1.ErrorPlot(nbins=nbins)
DS1.ErrorDistribution_xy(nbins=nbins, xlim=xlim, error=DS1.coloc_error, fit_data=True)
DS1.ErrorDistribution_r(nbins=nbins, xlim=xlim, error=DS1.coloc_error, mu=DS1.mu, fit_data=True)
#DS1.ErrorFOV()
#%% DS2 output
if DS2 is not None:
if not DS2.linked:
DS2.link_dataset(maxDistance=maxDistance)
DS2.Filter(pair_filter[1])
#DS2.ErrorPlot(nbins=nbins)
DS2.ErrorDistribution_xy(nbins=nbins, xlim=xlim, error=DS2.coloc_error)
DS2.ErrorDistribution_r(nbins=nbins, xlim=xlim, error=DS2.coloc_error, mu=DS2.mu)
#DS2.ErrorFOV()
#%% Image overview
fig,ax=DS1.show_channel(DS1.ch1.pos, ps=7)
DS1.show_channel(DS1.ch2.pos, ps=7, color='blue', fig=fig, ax=ax)
if False:
DS1.generate_channel(precision=DS1.pix_size)
DS1.plot_1channel()
if False and DS2 is not None:
DS2.generate_channel(precision=DS2.pix_size)
DS2.plot_1channel()
#%% model summary
DS1.model_summary()
if DS2 is not None: DS2.model_summary()