def run_shell(numx,
numy,
seed,
gip3x,
gip3y,
gcx=1000,
gcy=1000,
sparsity=0.002,
gc=1000,
gip3=5,
pathway="Fast",
save_interval=100,
density=0.1,
total_prop_time=0.1,
save_dir="../results/data/calcium",
**kargs):
# Check the existence of the saving directory
if not save_dir.endswith('/'):
save_dir += '/'
if not os.path.isdir(save_dir):
raise FileExistsError("Target directory " + save_dir +
" does not exist. ")
# Build layers
smc = SMC(**kargs)
ectoderm = Layer(smc,
numx=numx,
numy=numy,
gip3x=gip3x,
gip3y=gip3y,
gcx=gcx,
gcy=gcy,
save_interval=save_interval)
endoderm = Layer(smc,
numx=numx,
numy=numy,
gip3x=gip3x,
gip3y=gip3y,
gcx=gcx,
gcy=gcy,
save_interval=save_interval)
# Define stimulation times
stims_fast = [0.0, 4.4, 7.1, 9.3, 11.2, 13.2, 15.7, 18.4, 21.2, 24.7]
# Set stimulation patterns
# ectoderm.set_stim_pattern("fast", xmin=0, xmax=-1, ymin=0, ymax=-1,
# stim_times=stims_fast, randomnum=num_neurons,
# neighborsize=1)
delay = total_prop_time / numy
for row in range(numy):
for col in range(numx):
isneuron = bool(random.random() < density)
if isneuron:
ectoderm.set_stim_pattern(
"fast",
xmin=col,
xmax=min(col + 1, numx),
ymin=row,
ymax=min(row + 1, numy),
stim_times=[x + row * delay for x in stims_fast])
# Build shell
shell = Shell(ectoderm, endoderm, seed, sparsity, gc, gip3)
# Run the model
sol = shell.run()
sol = pd.DataFrame(sol)
# Generate filename and corresponding metadata
filename = datetime.datetime.now().strftime("%Y-%m-%d-%H-%M-%S-%f")
filemeta = "numx=" + str(numx) + ","
filemeta += "numy=" + str(numy) + ","
filemeta += "gc=" + str(gc) + ","
filemeta += "gip3=" + str(gip3) + ","
filemeta += "gip3x=" + str(gip3x) + ","
filemeta += "gip3y=" + str(gip3y) + ","
filemeta += "sparsity=" + str(sparsity) + ","
filemeta += "density=" + str(density) + ","
filemeta += "total_prop_time=" + str(total_prop_time) + ","
filemeta += "seed=" + str(seed)
for key in kargs:
filemeta += "," + key + '=' + str(kargs[key])
filemeta += ",shell"
# Save the results
sol.to_hdf(save_dir + filename + '.h5', 'calcium')
# Document the metadata
with open("../results/data/meta.txt", "a+") as metafile:
metafile.write(filename + " " + filemeta + '\n')
def run_shell(numx, numy, seed, gip3x, gip3y, gcx, gcy, sparsity, gc, gip3,
save_interval, save_dir, randomnum, **kargs):
# Check the existence of the saving directory
if not save_dir.endswith('/'):
save_dir += '/'
if not os.path.isdir(save_dir):
raise FileExistsError("Target directory " + save_dir +
" does not exist. ")
# Build layers
smc = SMC(**kargs)
ectoderm = Layer(smc,
numx=numx,
numy=numy,
gip3x=gip3x,
gip3y=gip3y,
gcx=gcx,
gcy=gcy,
save_interval=save_interval)
endoderm = Layer(smc,
numx=numx,
numy=numy,
gip3x=gip3x,
gip3y=gip3y,
gcx=gcx,
gcy=gcy,
save_interval=save_interval)
# Define stimulation times
stims_fast = []
stims_slow1 = [10]
# stims_slow2 = [15]
# stims_slow3 = [20]
# Set stimulation patterns
ectoderm.set_stim_pattern("slow",
xmin=0,
xmax=-1,
ymin=0,
ymax=-1,
stim_times=stims_slow1,
randomnum=randomnum)
# ectoderm.set_stim_pattern("slow", xmin=0, xmax=-1, ymin=0, ymax=-1,
# stim_times=stims_slow2, randomnum=randomnum)
# ectoderm.set_stim_pattern("slow", xmin=0, xmax=-1, ymin=0, ymax=-1,
# stim_times=stims_slow3, randomnum=randomnum)
# Build shell
shell = Shell(ectoderm, endoderm, seed, sparsity, gc, gip3)
# Run the model
sol = shell.run()
sol = pd.DataFrame(sol)
# Generate filename and corresponding metadata
filename = datetime.datetime.now().strftime("%Y-%m-%d-%H-%M-%S-%f")
filemeta = "numx=" + str(numx) + ","
filemeta += "numy=" + str(numy) + ","
filemeta += "gip3x=" + str(gip3x) + ","
filemeta += "gip3y=" + str(gip3y) + ","
filemeta += "randomnum=" + str(randomnum)
for key in kargs:
filemeta += "," + key + '=' + str(kargs[key])
filemeta += ",slowwave"
# Save the results
sol.to_hdf(save_dir + filename + '.h5', 'calcium')
# Document the metadata
with open("../results/data/meta.txt", "a+") as metafile:
metafile.write(filename + " " + filemeta + '\n')
def run_shell(numx,
numy,
seed,
gip3x,
gip3y,
gcx=1000,
gcy=1000,
sparsity=0.002,
gc=1000,
gip3=5,
pathway="Fast",
save_interval=100,
num_neurons=5,
save_dir="../results/data/calcium",
**kargs):
# Check the existence of the saving directory
if not save_dir.endswith('/'):
save_dir += '/'
if not os.path.isdir(save_dir):
raise FileExistsError("Target directory " + save_dir +
" does not exist. ")
# Build layers
smc = SMC(**kargs)
ectoderm = Layer(smc,
numx=numx,
numy=numy,
gip3x=gip3x,
gip3y=gip3y,
gcx=gcx,
gcy=gcy,
save_interval=save_interval)
endoderm = Layer(smc,
numx=numx,
numy=numy,
gip3x=gip3x,
gip3y=gip3y,
gcx=gcx,
gcy=gcy,
save_interval=save_interval)
# Define stimulation times
stims_fast = [0.0, 5.0, 10.0]
# Set stimulation patterns
# ectoderm.set_stim_pattern("fast", xmin=0, xmax=numx, ymin=0, ymax=1,
# stim_times=stims_fast)
ectoderm.set_stim_pattern("fast",
xmin=0,
xmax=-1,
ymin=0,
ymax=-1,
stim_times=stims_fast,
randomnum=num_neurons,
neighborsize=1)
# Build shell
shell = Shell(ectoderm, endoderm, seed, sparsity, gc, gip3)
# Run the model
sol = shell.run()
sol = pd.DataFrame(sol)
# Generate filename and corresponding metadata
filename = datetime.datetime.now().strftime("%Y-%m-%d-%H-%M-%S-%f")
filemeta = "numx=" + str(numx) + ","
filemeta += "numy=" + str(numy) + ","
filemeta += "gc=" + str(gc) + ","
filemeta += "gip3=" + str(gip3) + ","
filemeta += "gip3x=" + str(gip3x) + ","
filemeta += "gip3y=" + str(gip3y) + ","
filemeta += "sparsity=" + str(sparsity) + ","
filemeta += "num_neurons=" + str(num_neurons) + ","
filemeta += "seed=" + str(seed)
for key in kargs:
filemeta += "," + key + '=' + str(kargs[key])
filemeta += ",shell"
# Save the results
sol.to_hdf(save_dir + filename + '.h5', 'calcium')
# Document the metadata
with open("../results/data/meta.txt", "a+") as metafile:
metafile.write(filename + " " + filemeta + '\n')
def run_layer(numx,
numy,
gip3x,
gip3y,
gcx=1000,
gcy=1000,
pathway="Both",
save_interval=100,
save_dir="../results/data/calcium",
**kargs):
"Run the Layer"
# Check the existence of the saving directory
if not save_dir.endswith('/'):
save_dir += '/'
if not os.path.isdir(save_dir):
raise FileExistsError("Target directory " + save_dir +
" does not exist. ")
# Initialize model
smc = SMC(**kargs)
layer = Layer(smc,
numx=numx,
numy=numy,
gip3x=gip3x,
gip3y=gip3y,
gcx=gcx,
gcy=gcy,
save_interval=save_interval)
# Define stimulation times
stims_fast = [
0, 4.4, 7.1, 9.3, 11.2, 13.2, 15.7, 18.4, 21.2, 24.7, 100, 105.2,
108.2, 110.6, 112.8, 115, 117.3, 119.4, 121.9, 125.1, 129.5, 134.3,
200, 205.7, 208.8, 211.6, 213.8, 216.1, 218.3, 221, 224.2, 229, 235.4
]
# stims_slow1 = [114]
stims_slow1 = [14]
stims_slow2 = [214]
# Set stimulation patterns
if pathway == "Both":
layer.set_stim_pattern("fast",
xmin=0,
xmax=numx,
ymin=0,
ymax=1,
stim_times=stims_fast)
layer.set_stim_pattern("slow",
xmin=90,
xmax=110,
ymin=0,
ymax=10,
stim_times=stims_slow1)
layer.set_stim_pattern("slow",
xmin=0,
xmax=10,
ymin=0,
ymax=10,
stim_times=stims_slow2)
layer.set_stim_pattern("slow",
xmin=numx - 10,
xmax=numx,
ymin=0,
ymax=10,
stim_times=stims_slow2)
elif pathway == "Fast":
layer.set_stim_pattern("fast",
xmin=0,
xmax=numx,
ymin=0,
ymax=1,
stim_times=[0])
elif pathway == "Slow":
# layer.set_stim_pattern("slow", xmin=13, xmax=17, ymin=0, ymax=4,
# stim_times=[0])
layer.set_stim_pattern("slow",
xmin=0,
xmax=-1,
ymin=0,
ymax=-1,
stim_times=[0],
randomnum=1,
neighborsize=2)
# Run the model
sol = layer.run()
sol = pd.DataFrame(sol)
# Generate filename and corresponding metadata
filename = datetime.datetime.now().strftime("%Y-%m-%d-%H-%M-%S-%f")
filemeta = "numx=" + str(numx) + ","
filemeta += "numy=" + str(numy) + ","
filemeta += "gip3x=" + str(gip3x) + ","
filemeta += "gip3y=" + str(gip3y) + ","
filemeta += "gcx=" + str(gcx)
for key in kargs:
filemeta += "," + key + '=' + str(kargs[key])
if pathway == "Fast":
filemeta += ",endo"
# Save the results
sol.to_hdf(save_dir + filename + '.h5', 'calcium')
# Document the metadata
with open("../results/data/meta.txt", "a+") as metafile:
metafile.write(filename + " " + filemeta + '\n')
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
from tqdm import tqdm
import sys
from hydramuscle.model.smc import SMC
from hydramuscle.model.layer import Layer
from hydramuscle.model.shell import Shell
from hydramuscle.model.fluo_encoder import FluoEncoder
from hydramuscle.utils import utils
if __name__ == "__main__":
smc = SMC(k_ipr=0.2, k_deg=0.05, s0=100, d=20e-4, T=50, dt=0.0002)
ectoderm = Layer(smc,
numx=30,
numy=60,
gip3x=0.1,
gip3y=2,
gcx=1000,
gcy=1000,
save_interval=100,
active_v_beta=1)
endoderm = Layer(smc,
numx=30,
numy=60,
gip3x=0.1,
gip3y=2,
gcx=1000,
gcy=1000,
save_interval=100,
def run_shell(numx, numy, seed, gip3x, gip3y, gcx=1000, gcy=1000, sparsity=0.002, gc=1000, gip3=5, pathway="Both",
save_interval=100, active_v_beta=1, save_dir="../results/data/calcium", **kargs):
# Check the existence of the saving directory
if not save_dir.endswith('/'):
save_dir += '/'
if not os.path.isdir(save_dir):
raise FileExistsError("Target directory " + save_dir + " does not exist. ")
# Build layers
smc = SMC(**kargs)
ectoderm = Layer(smc, numx=numx, numy=numy, gip3x=gip3x, gip3y=gip3y,
gcx=gcx, gcy=gcy, save_interval=save_interval, active_v_beta=active_v_beta)
endoderm = Layer(smc, numx=numx, numy=numy, gip3x=gip3x, gip3y=gip3y,
gcx=gcx, gcy=gcy, save_interval=save_interval, active_v_beta=active_v_beta)
# Define stimulation times
# init1 = 100
# init2 = 208 # 184.3
# init3 = 339.75 # 284.7
# init4 = 462.25 # 376.4
# stims_fast = [0.0, 9.5, 15.5, 18.5, 22.25, 25.75, 29.0, 32.5, 36.0, 43.25,
# 100.0, 109.5, 115.5, 118.5, 122.25, 125.75, 129.0, 132.5, 136.0, 143.25,
# 208.0, 209.75, 217.75, 223.75, 227.25, 230.25, 233.75, 237.5, 242.0, 249.75,
# 339.75, 347.25, 352.5, 356.25, 359.75, 363.0, 366.25, 370.0, 375.25,
# 462.25, 471.5, 476.75, 480.5, 483.75, 486.75, 490.25, 493.75, 497.5, 502.75]
# stims_fast = [0.0, 4.4, 7.1, 9.3, 11.2, 13.2, 15.7, 18.4, 21.2, 24.7,
# 100.0, 104.4, 107.1, 109.3, 111.2, 113.2, 115.7, 118.4, 121.2, 124.7,
# 184.3, 188.0, 189.3, 190.4, 193.4, 195.0, 197.1, 199.4, 201.6, 204.8, 208.9, 215.8, 219.3, 222.2, 224.7, 227.6, 232.1,
# 284.7, 289.9, 292.9, 295.3, 297.5, 299.7, 302.0, 304.1, 306.6, 309.8, 314.2, 319.0,
# 376.4, 382.1, 385.2, 388.0, 390.2, 392.5, 394.7, 397.4, 400.6, 405.4, 411.8]
stims_fast = [0.0, 5.2, 9.0, 12.1, 15.8, 19.6, 23.5, 27.0, 31.7,
99.3, 105.6, 110.4, 115.3, 119.9, 124.8, 128.6, 132.9,
249.5, 256.5, 262.0, 267.0, 271.5, 276.6, 283.2, 288.7, 293.4, 298.8, 305.6,
433.5, 440.0, 456.0, 461.7, 466.5, 471.0, 477.3, 483.1, 489.8]
# stims_fast = []
# stims_slow1 = [20, init1+20, init3+20]
# stims_slow2 = [init2+20, init4+20]
stims_slow1 = []
stims_slow2 = []
# Set stimulation patterns
if pathway == "Both":
ectoderm.set_stim_pattern("fast", xmin=0, xmax=numx, ymin=0, ymax=1, stim_times=stims_fast)
ectoderm.set_stim_pattern("slow", xmin=13, xmax=17, ymin=0, ymax=4,
stim_times=stims_slow1)
ectoderm.set_stim_pattern("slow", xmin=0, xmax=2, ymin=0, ymax=4,
stim_times=stims_slow2)
ectoderm.set_stim_pattern("slow", xmin=numx-2, xmax=numx, ymin=0, ymax=4,
stim_times=stims_slow2)
elif pathway == "Fast":
ectoderm.set_stim_pattern("fast", xmin=0, xmax=numx, ymin=0, ymax=1, stim_times=stims_fast)
elif pathway == "Slow":
ectoderm.set_stim_pattern("slow", xmin=13, xmax=17, ymin=0, ymax=4,
stim_times=stims_slow1)
# Build shell
shell = Shell(ectoderm, endoderm, seed, sparsity, gc, gip3)
# Run the model
sol = shell.run()
sol = pd.DataFrame(sol)
# Generate filename and corresponding metadata
filename = datetime.datetime.now().strftime("%Y-%m-%d-%H-%M-%S-%f")
filemeta = "numx=" + str(numx) + ","
filemeta += "numy=" + str(numy) + ","
filemeta += "gc=" + str(gc) + ","
filemeta += "gip3=" + str(gip3) + ","
filemeta += "gip3x=" + str(gip3x) + ","
filemeta += "gip3y=" + str(gip3y) + ","
filemeta += "sparsity=" + str(sparsity) + ","
filemeta += "active_v_beta=" + str(active_v_beta) + ","
filemeta += "seed=" + str(seed)
for key in kargs:
filemeta += "," + key + '=' + str(kargs[key])
filemeta += ",shell"
# Save the results
sol.to_hdf(save_dir + filename + '.h5', 'calcium')
# Document the metadata
with open("../results/data/meta.txt", "a+") as metafile:
metafile.write(filename + " " + filemeta + '\n')