New plots for swp2 output files
tforest
parent
f79d48bf5b
commit
463b9e63f9
246
swp2.py
246
swp2.py
|
|
@ -1,7 +1,8 @@
|
|||
import matplotlib.pyplot as plt
|
||||
import os
|
||||
import numpy as np
|
||||
|
||||
def return_x_y_from_stwp_theta_file(stwp_theta_file, breaks, mu, tgen):
|
||||
def return_x_y_from_stwp_theta_file(stwp_theta_file, breaks, mu, tgen, relative_theta_scale = False):
|
||||
with open(stwp_theta_file, "r") as swp_file:
|
||||
# Read the first line
|
||||
line = swp_file.readline()
|
||||
|
|
@ -53,18 +54,27 @@ def return_x_y_from_stwp_theta_file(stwp_theta_file, breaks, mu, tgen):
|
|||
#t =
|
||||
if len(group.split(',')) == 1:
|
||||
k = i
|
||||
t[i] += ((theta_L[group_nb] ) / (k*(k-1)) * tgen) / mu
|
||||
if relative_theta_scale:
|
||||
t[i] += ((theta_L[group_nb] ) / (k*(k-1)))
|
||||
else:
|
||||
t[i] += ((theta_L[group_nb] ) / (k*(k-1)) * tgen) / mu
|
||||
else:
|
||||
for k in range(j, i-1, -1 ):
|
||||
t[i] += ((theta_L[group_nb] ) / (k*(k-1)) * tgen) / mu
|
||||
if relative_theta_scale:
|
||||
t[i] += ((theta_L[group_nb] ) / (k*(k-1)))
|
||||
else:
|
||||
t[i] += ((theta_L[group_nb] ) / (k*(k-1)) * tgen) / mu
|
||||
# we add the cumulative times at the end
|
||||
t[i] += sum_t
|
||||
sum_t = t[i]
|
||||
# build the y axis (sizes)
|
||||
y = []
|
||||
for theta in theta_L:
|
||||
# with size N = theta/4mu
|
||||
size = theta / (4*mu)
|
||||
if relative_theta_scale:
|
||||
size = theta
|
||||
else:
|
||||
# with size N = theta/4mu
|
||||
size = theta / (4*mu)
|
||||
y.append(size)
|
||||
y.append(size)
|
||||
# build the time x axis
|
||||
|
|
@ -73,19 +83,64 @@ def return_x_y_from_stwp_theta_file(stwp_theta_file, breaks, mu, tgen):
|
|||
x.append(list(t.values())[time])
|
||||
x.append(list(t.values())[time])
|
||||
x.append(list(t.values())[len(t.values())-1])
|
||||
# if relative_theta_scale:
|
||||
# # rescale
|
||||
# #N0 = y[0]
|
||||
# # for i in range(len(y)):
|
||||
# # # divide by N0
|
||||
# # y[i] = y[i]/N0
|
||||
# # x[i] = x[i]/N0
|
||||
return x,y,likelihood
|
||||
|
||||
def plot_3epochs_thetafolder(folder_path, mu, tgen, breaks = 2, title = "Title"):
|
||||
def return_x_y_from_stwp_theta_file_as_is(stwp_theta_file, breaks, mu, tgen, relative_theta_scale = False):
|
||||
with open(stwp_theta_file, "r") as swp_file:
|
||||
# Read the first line
|
||||
line = swp_file.readline()
|
||||
L = float(line.split()[2])
|
||||
# Process lines until the end of the file
|
||||
while line:
|
||||
# check at each line
|
||||
if line.startswith("dim") :
|
||||
dim = int(line.split()[1])
|
||||
if dim == breaks+1:
|
||||
likelihood = line.split()[5]
|
||||
groups = line.split()[6:6+dim]
|
||||
theta_site = line.split()[6+dim:6+dim+1+dim]
|
||||
elif dim < breaks+1:
|
||||
line = swp_file.readline()
|
||||
continue
|
||||
elif dim > breaks+1:
|
||||
break
|
||||
#return 0,0,0
|
||||
# Read the next line
|
||||
line = swp_file.readline()
|
||||
#### END of parsing
|
||||
# quit this file if the number of dimensions is incorrect
|
||||
if dim < breaks+1:
|
||||
return 0,0,0
|
||||
# get n, the last bin of the last group
|
||||
# revert the list of groups as the most recent times correspond
|
||||
# to the closest and last leafs of the coal. tree.
|
||||
groups = groups[::-1]
|
||||
theta_site = theta_site[::-1]
|
||||
|
||||
thetas = {}
|
||||
|
||||
for i in range(len(groups)):
|
||||
groups[i] = groups[i].split(',')
|
||||
#print(groups[i], len(groups[i]))
|
||||
thetas[i] = [float(theta_site[i]), groups[i], likelihood]
|
||||
return thetas
|
||||
|
||||
def plot_k_epochs_thetafolder(folder_path, mu, tgen, breaks = 2, title = "Title", theta_scale = True):
|
||||
scenari = {}
|
||||
cpt = 0
|
||||
|
||||
for file_name in os.listdir(folder_path):
|
||||
if os.path.isfile(os.path.join(folder_path, file_name)):
|
||||
# Perform actions on each file
|
||||
x,y,likelihood = return_x_y_from_stwp_theta_file(folder_path+file_name, breaks = breaks,
|
||||
tgen = tgen,
|
||||
mu = mu)
|
||||
mu = mu, relative_theta_scale = theta_scale)
|
||||
if x == 0 or y == 0:
|
||||
continue
|
||||
cpt +=1
|
||||
|
|
@ -93,6 +148,7 @@ def plot_3epochs_thetafolder(folder_path, mu, tgen, breaks = 2, title = "Title")
|
|||
print("\n*******\n"+title+"\n--------\n"+"mu="+str(mu)+"\ntgen="+str(tgen)+"\nbreaks="+str(breaks)+"\n*******\n")
|
||||
print(cpt, "theta file(s) have been scanned.")
|
||||
# sort starting by the smallest -log(Likelihood)
|
||||
print(scenari)
|
||||
best10_scenari = (sorted(list(scenari.keys())))[:10]
|
||||
print("10 greatest Likelihoods", best10_scenari)
|
||||
greatest_likelihood = best10_scenari[0]
|
||||
|
|
@ -100,21 +156,181 @@ def plot_3epochs_thetafolder(folder_path, mu, tgen, breaks = 2, title = "Title")
|
|||
my_dpi = 300
|
||||
plt.figure(figsize=(5000/my_dpi, 2800/my_dpi), dpi=my_dpi)
|
||||
plt.plot(x, y, 'r-', lw=2, label = 'Lik='+greatest_likelihood)
|
||||
plt.yscale('log')
|
||||
#plt.xscale('log')
|
||||
plt.grid(True,which="both", linestyle='--')
|
||||
plt.xlim(1e-3, 1)
|
||||
plt.ylim(0, 10)
|
||||
#plt.yscale('log')
|
||||
plt.xscale('log')
|
||||
plt.grid(True,which="both", linestyle='--', alpha = 0.3)
|
||||
|
||||
for scenario in best10_scenari[1:]:
|
||||
x,y = scenari[scenario]
|
||||
#print("\n---- Lik:",scenario,"\n\nt=", x,"\n\nN=",y, "\n\n")
|
||||
plt.plot(x, y, '--', lw=1, label = 'Lik='+scenario)
|
||||
plt.ylabel("Individuals (N)")
|
||||
plt.xlabel("Time (years)")
|
||||
if theta_scale:
|
||||
plt.xlabel("Coal. time")
|
||||
plt.ylabel("Pop. size scaled by N0")
|
||||
recent_scale_lower_bound = y[0] * 0.01
|
||||
recent_scale_upper_bound = y[0] * 0.1
|
||||
plt.axvline(x=recent_scale_lower_bound)
|
||||
plt.axvline(x=recent_scale_upper_bound)
|
||||
else:
|
||||
# years
|
||||
plt.xlabel("Time (years)")
|
||||
plt.ylabel("Individuals (N)")
|
||||
plt.legend(loc='upper right')
|
||||
plt.title(title)
|
||||
#plt.gcf().set_size(1000, 500)
|
||||
plt.savefig(title+'_b'+str(breaks)+'.pdf')
|
||||
|
||||
def plot_all_epochs_thetafolder(folder_path, mu, tgen, title = "Title", theta_scale = True):
|
||||
#scenari = {}
|
||||
cpt = 0
|
||||
epochs = {}
|
||||
for file_name in os.listdir(folder_path):
|
||||
breaks = 0
|
||||
cpt +=1
|
||||
if os.path.isfile(os.path.join(folder_path, file_name)):
|
||||
x, y, likelihood = return_x_y_from_stwp_theta_file(folder_path+file_name, breaks = breaks,
|
||||
tgen = tgen,
|
||||
mu = mu, relative_theta_scale = theta_scale)
|
||||
while not (x == 0 and y == 0):
|
||||
if breaks not in epochs.keys():
|
||||
epochs[breaks] = {}
|
||||
epochs[breaks][likelihood] = x,y
|
||||
breaks += 1
|
||||
x,y,likelihood = return_x_y_from_stwp_theta_file(folder_path+file_name, breaks = breaks,
|
||||
tgen = tgen,
|
||||
mu = mu, relative_theta_scale = theta_scale)
|
||||
print("\n*******\n"+title+"\n--------\n"+"mu="+str(mu)+"\ntgen="+str(tgen)+"\nbreaks="+str(breaks)+"\n*******\n")
|
||||
print(cpt, "theta file(s) have been scanned.")
|
||||
|
||||
# intialize figure
|
||||
my_dpi = 300
|
||||
plt.figure(figsize=(5000/my_dpi, 2800/my_dpi), dpi=my_dpi)
|
||||
plt.xlim(1e-3, 1)
|
||||
#plt.ylim(0, 10)
|
||||
#plt.yscale('log')
|
||||
plt.xscale('log')
|
||||
plt.grid(True,which="both", linestyle='--', alpha = 0.3)
|
||||
brkpt_lik = []
|
||||
for epoch, scenari in epochs.items():
|
||||
# sort starting by the smallest -log(Likelihood)
|
||||
best10_scenari = (sorted(list(scenari.keys())))[:10]
|
||||
greatest_likelihood = best10_scenari[0]
|
||||
# store the tuple breakpoints and likelihood for later plot
|
||||
brkpt_lik.append((epoch, greatest_likelihood))
|
||||
x, y = scenari[greatest_likelihood]
|
||||
#without breakpoint
|
||||
if epoch == 0:
|
||||
# do something with the theta without bp and skip the plotting
|
||||
N0 = y[0]
|
||||
#continue
|
||||
for i in range(len(y)):
|
||||
# divide by N0
|
||||
y[i] = y[i]/N0
|
||||
x[i] = x[i]/N0
|
||||
plt.plot(x, y, '-', alpha=0.75, lw=2, label = str(epoch)+' BrkPt | Lik='+greatest_likelihood)
|
||||
if theta_scale:
|
||||
plt.xlabel("Coal. time")
|
||||
plt.ylabel("Pop. size scaled by N0")
|
||||
recent_scale_lower_bound = 0.01
|
||||
recent_scale_upper_bound = 0.1
|
||||
#print(recent_scale_lower_bound, recent_scale_upper_bound)
|
||||
plt.axvline(x=recent_scale_lower_bound)
|
||||
plt.axvline(x=recent_scale_upper_bound)
|
||||
else:
|
||||
# years
|
||||
plt.xlabel("Time (years)")
|
||||
plt.ylabel("Individuals (N)")
|
||||
plt.xlim(1e-5, 1)
|
||||
plt.legend(loc='upper right')
|
||||
plt.title(title)
|
||||
plt.savefig(title+'_b'+str(breaks)+'.pdf')
|
||||
# plot likelihood against nb of breakpoints
|
||||
plt.figure(figsize=(5000/my_dpi, 2800/my_dpi), dpi=my_dpi)
|
||||
plt.rcParams['font.size'] = '18'
|
||||
AIC = 2*(len(brkpt_lik)+1)+2*np.array(brkpt_lik)[:, 1].astype(float)
|
||||
plt.plot(np.array(brkpt_lik)[:, 0], AIC, 'o', linestyle = "dotted", lw=2)
|
||||
plt.axhline(y=106)
|
||||
plt.yscale('log')
|
||||
plt.xlabel("# breakpoints", fontsize=20)
|
||||
plt.ylabel("$-\log\mathcal{L}$")
|
||||
#plt.legend(loc='upper right')
|
||||
plt.title(title)
|
||||
plt.savefig(title+'_Breakpts_Likelihood.pdf')
|
||||
|
||||
def plot_test_theta(folder_path, mu, tgen, title = "Title", theta_scale = True, breaks_max = 6):
|
||||
cpt = 0
|
||||
epochs = {}
|
||||
for file_name in os.listdir(folder_path):
|
||||
cpt +=1
|
||||
if os.path.isfile(os.path.join(folder_path, file_name)):
|
||||
for k in range(breaks_max):
|
||||
thetas = return_x_y_from_stwp_theta_file_as_is(folder_path+file_name, breaks = k,
|
||||
tgen = tgen,
|
||||
mu = mu, relative_theta_scale = theta_scale)
|
||||
if thetas[0] == 0:
|
||||
continue
|
||||
epochs[k] = thetas
|
||||
print("\n*******\n"+title+"\n--------\n"+"mu="+str(mu)+"\ntgen="+str(tgen)+"\nbreaks="+str(k)+"\n*******\n")
|
||||
print(cpt, "theta file(s) have been scanned.")
|
||||
|
||||
# intialize figure
|
||||
my_dpi = 300
|
||||
plt.figure(figsize=(5000/my_dpi, 2800/my_dpi), dpi=my_dpi)
|
||||
for epoch, theta in epochs.items():
|
||||
groups = np.array(list(theta.values()), dtype=object)[:, 1].tolist()
|
||||
x = []
|
||||
y = []
|
||||
thetas = np.array(list(theta.values()), dtype=object)[:, 0]
|
||||
for i,group in enumerate(groups):
|
||||
x += group[::-1]
|
||||
y += list(np.repeat(thetas[i], len(group)))
|
||||
if epoch == 0:
|
||||
N0 = y[0]
|
||||
for i in range(len(y)):
|
||||
y[i] = y[i]/N0
|
||||
plt.plot(x, y, 'o', linestyle="dotted", alpha=0.75, lw=2, label = str(epoch)+' brks')
|
||||
plt.xlabel("# breaks")
|
||||
plt.ylabel("theta")
|
||||
plt.legend(loc='upper right')
|
||||
plt.savefig(title+'_test'+str(k)+'.pdf')
|
||||
# fig 2
|
||||
plt.figure(figsize=(5000/my_dpi, 2800/my_dpi), dpi=my_dpi)
|
||||
for epoch, theta in epochs.items():
|
||||
groups = np.array(list(theta.values()), dtype=object)[:, 1].tolist()
|
||||
x = []
|
||||
y = []
|
||||
thetas = np.array(list(theta.values()), dtype=object)[:, 0]
|
||||
for i,group in enumerate(groups):
|
||||
x += group[::-1]
|
||||
y += list(np.repeat(thetas[i], len(group)))
|
||||
if epoch == 0:
|
||||
N0 = y[0]
|
||||
for i in range(len(y)):
|
||||
y[i] = y[i]/N0
|
||||
#
|
||||
x_2 = []
|
||||
T = 0
|
||||
# k allant de de 14 à 2
|
||||
for i in range(len(x)):
|
||||
x[i] = int(x[i])
|
||||
#print(x[2])
|
||||
for i in range(0, len(x)):
|
||||
k = x[i]
|
||||
#print(k, y[k-2])
|
||||
|
||||
#theta_k = y[k] / (k*(k-1))
|
||||
T += y[i] / (x[i]*(x[i]-1))
|
||||
x_2.append(T)
|
||||
|
||||
plt.plot(x_2, y, 'o', linestyle="dotted", alpha=0.75, lw=2, label = str(epoch)+' brks')
|
||||
plt.xscale('log')
|
||||
plt.xlabel("# breaks")
|
||||
plt.ylabel("theta")
|
||||
plt.legend(loc='upper right')
|
||||
plt.savefig(title+'_test'+str(k)+'.pdf')
|
||||
#
|
||||
|
||||
if __name__ == "__main__":
|
||||
|
||||
if len(sys.argv) != 4:
|
||||
|
|
@ -125,4 +341,4 @@ if __name__ == "__main__":
|
|||
mu = sys.argv[2]
|
||||
tgen = sys.argv[3]
|
||||
|
||||
plot_3epochs_thetafolder(folder_path, mu, tgen)
|
||||
plot_all_epochs_thetafolder(folder_path, mu, tgen)
|
||||
|
|
|
|||
Loading…
Reference in New Issue