-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathrun_training.py
318 lines (234 loc) · 9.74 KB
/
run_training.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
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
#!/usr/bin/env python
# coding: utf-8
# In[15]:
import numpy as np
from pprint import pprint
import sys, os
import pickle as pkl
from collections import defaultdict
import logging
import argparse
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)
from neuron import h
from neuron.units import ms, mV
h.nrnmpi_init()
import time
def get_cell_population_spikes(c,pop_id):
spike_times_dict = c.get_cell_spikes(pop_id)
return spike_times_dict
def get_ext_population_spikes(c,pop_id):
spike_vec_dict = defaultdict(list)
spike_times_vec = c.external_spike_time_recs[pop_id]
spike_gids_vec = c.external_spike_gid_recs[pop_id]
for spike_t, spike_gid in zip(spike_times_vec, spike_gids_vec):
spike_gid = int(spike_gid)
spike_vec_dict[spike_gid].append(spike_t)
return spike_vec_dict
def get_population_voltages(c,pop_id,rec_dt=0.1):
v_vec_dict = {}
for cid, cell in c.neurons[pop_id].items():
v_vec = h.Vector()
try:
v_vec.record(cell.axon(0.5)._ref_v, rec_dt)
except:
v_vec.record(cell.soma(0.5)._ref_v, rec_dt)
gid = c.ctype_offsets[pop_id] + cid
v_vec_dict[gid] = v_vec
return v_vec_dict
def pull_spike_times(population2info_dict):
spike_times = {}
gids = np.sort(list(population2info_dict.keys()))
for gid in gids:
gid_info = population2info_dict[gid]
if 'spike times' in gid_info:
spike_times[gid] = gid_info['spike times']
return spike_times
def main():
parser = argparse.ArgumentParser(
description="Run CA3 cue cell training."
)
parser.add_argument(
"--circuit-config",
required=True,
type=str,
help="Name of circuit configuration file. ",
)
parser.add_argument(
"--arena-config",
required=True,
type=str,
help="Name of arena configuration file. ",
)
parser.add_argument(
"--model-home",
required=False,
type=str,
help="Path to model home directory. ",
)
parser.add_argument(
"--data-prefix",
required=False,
default="data",
type=str,
help="Path to data files. ",
)
parser.add_argument(
"-c", "--config-id",
required=True,
type=str,
help="Configuration identifier. ",
)
parser.add_argument(
"-s", "--save-weights-every",
required=False,
type=int, default=-1,
help="Save weights every n laps. ",
)
args = parser.parse_args()
model_home = args.model_home
sys.path.append(os.path.join(model_home, 'utils'))
sys.path.append(os.path.join(model_home, 'cells'))
circuit_config_file = args.circuit_config
arena_config_file = args.arena_config
config_id = args.config_id
data_prefix = args.data_prefix
save_weights_every = args.save_weights_every
from SetupConnections import WiringDiagram, Arena
from NeuronCircuit import Circuit, save_v_vecs, save_netcon_data, save_spike_vecs
from simtime import SimTimeEvent
print('constructing model..')
sys.stdout.flush()
delay = 500.
dt = 0.1
pc = h.ParallelContext()
params_path = os.path.join(model_home, 'params')
arena_config_path = os.path.join(params_path, arena_config_file)
circuit_config_path = os.path.join(params_path, circuit_config_file)
ar = Arena(os.path.join(params_path, arena_config_path))
ar.generate_population_firing_rates()
ar.generate_cue_firing_rates('LEC', 1.0)
cued = True
fr = ar.cell_information['LEC']['cell info'][0]['firing rate']
edge = 12.5
lp = 1
arena_size = ar.params['Arena']['arena size']
bin_size = ar.params['Arena']['bin size']
mouse_speed = ar.params['Arena']['mouse speed']
nlaps = ar.params['Arena']['lap information']['nlaps']
arena_map = np.arange(0, 200,step=0.1)
cued_positions = np.linspace(edge, 200-edge, nlaps*lp)
random_cue_locs = np.arange(len(cued_positions))
if pc.id() == 0:
np.random.shuffle(random_cue_locs)
random_cue_locs = pc.py_broadcast(random_cue_locs, 0)
time_for_single_lap = arena_size / mouse_speed * 1000.
frs_all = []
for i in range(nlaps):
random_position = cued_positions[random_cue_locs[i]]
to_roll = int( ( 100. - random_position) / 0.1 )
fr_rolled = np.roll(fr, to_roll)
frs_all.append(fr_rolled)
frs_all = np.asarray(frs_all)
place_information = {'place ids': [0], 'place fracs': [0.80]}
diagram = WiringDiagram(circuit_config_path, place_information)
place_ids = diagram.place_information[0]['place']
cue_ids = diagram.place_information[0]['not place']
internal_kwargs = {}
internal_kwargs['place information'] = diagram.place_information
internal_kwargs['cue information'] = diagram.place_information
diagram.generate_internal_connectivity(**internal_kwargs)
external_kwargs = {}
external_kwargs['place information'] = diagram.place_information
external_kwargs['external place ids'] = [100, 101, 102]
external_kwargs['cue information'] = diagram.place_information
external_kwargs['external cue ids'] = [100, 101, 102]
diagram.generate_external_connectivity(ar.cell_information, **external_kwargs)
diagram.generate_septal_connectivity()
ar.generate_spike_times('MF', dt=dt, delay=delay)
ar.generate_spike_times('MEC', dt=dt, delay=delay)
ar.generate_spike_times('LEC', dt=dt, delay=delay, cued=cued)
ar.generate_spike_times('Background', dt=dt, delay=delay)
sys.stdout.flush()
mf_spike_times = pull_spike_times(ar.cell_information['MF']['cell info'])
mec_spike_times = pull_spike_times(ar.cell_information['MEC']['cell info'])
lec_spike_times = pull_spike_times(ar.cell_information['LEC']['cell info'])
bk_spike_times = pull_spike_times(ar.cell_information['Background']['cell info'])
print('constructing circuit..')
sys.stdout.flush()
circuit = Circuit(params_prefix=params_path,
params_filename=circuit_config_file,
arena_params_filename=arena_config_file,
internal_pop2id=diagram.pop2id,
external_pop2id=diagram.external_pop2id,
external_spike_times = {100: mf_spike_times,
101: mec_spike_times,
102: lec_spike_times,
103: bk_spike_times})
print('building cells..')
sys.stdout.flush()
circuit.build_cells()
circuit.build_internal_netcons(diagram.internal_adj_matrices, diagram.internal_ws)
circuit.build_external_netcons(100, diagram.external_adj_matrices[100], diagram.external_ws[100])
circuit.build_external_netcons(101, diagram.external_adj_matrices[101], diagram.external_ws[101])
circuit.build_external_netcons(102, diagram.external_adj_matrices[102], diagram.external_ws[102])
circuit.build_external_netcons(103, diagram.external_adj_matrices[103], diagram.external_ws[103])
#circuit.record_lfp([0,1])
#circuit.build_septal_netcons(diagram.septal_adj_matrices)
pc = circuit.pc
exc_v_vecs = get_population_voltages(circuit, 0)
#pvbc_v_vecs = get_population_voltages(circuit, 1)
# aac_v_vecs = get_population_voltages(2)
# bis_v_vecs = get_population_voltages(3)
# olm_v_vecs = get_population_voltages(4)
# isccr_v_vecs = get_population_voltages(5)
# iscck_v_vecs = get_population_voltages(6)
t_vec = h.Vector() # Time stamp vector
t_vec.record(h._ref_t)
tic = time.time()
h.dt = 0.025
h.celsius = 37.
if save_weights_every < 0:
save_weights_every = nlaps
t = h.Vector().record(h._ref_t)
simtime = SimTimeEvent(pc, time_for_single_lap * nlaps + delay, 8.0, 10, 0)
mindelay = pc.set_maxstep(10 * ms)
h.finitialize(-65 * mV)
for ilap in range(nlaps):
h.tstop = time_for_single_lap*(ilap + 1) + delay
if pc.id() == 0:
print(f'starting simulation for lap {ilap}/{nlaps} until {h.tstop} ms..')
sys.stdout.flush()
pc.set_maxstep(10 * ms)
pc.psolve(h.tstop - mindelay)
elapsed = time.time() - tic
pc.barrier()
if (ilap + 1) % save_weights_every == 0:
save_netcon_data(pc, circuit,
os.path.join(data_prefix, f"{config_id}-cue-ee-ei-nlaps-{(ilap+1)}-dt-zerodot1-scale-2-v1.npz"))
if pc.id() == 0:
print('simulation took %0.3f seconds' % elapsed)
ext_spikes_MF = get_ext_population_spikes(circuit, 100)
ext_spikes_MEC = get_ext_population_spikes(circuit, 101)
ext_spikes_LEC = get_ext_population_spikes(circuit, 102)
ext_spikes_Bk = get_ext_population_spikes(circuit, 103)
save_spike_vecs(pc,
os.path.join(data_prefix, f"ext_spikes_{config_id}-cue-ee-ei-nlaps-{nlaps}"),
ext_spikes_MF,
ext_spikes_MEC,
ext_spikes_LEC,
ext_spikes_Bk)
cell_spikes_PC = get_cell_population_spikes(circuit,0)
cell_spikes_PVBC = get_cell_population_spikes(circuit,1)
save_spike_vecs(pc,
os.path.join(data_prefix, f"cell_spikes_{config_id}-cue-ee-ei-nlaps-{nlaps}"),
cell_spikes_PC,
cell_spikes_PVBC)
save_v_vecs(pc,
os.path.join(data_prefix, f"v_vecs_{config_id}-cue-ee-ei-nlaps-{nlaps}"),
exc_v_vecs)
pc.runworker()
pc.done()
h.quit()
if __name__ == "__main__":
main()