def SI(self):
p_list = []
t_list = []
infected_list = []
for p in self.frange(0, 1, 0.1):
t, num_infected = self.infect(p)
p_list.append(p)
t_list.append(t)
infected_list.append(
float(num_infected) / self.network.number_of_nodes())
plt.title("Infected Length vs Probability of Infection")
plt.plot(p_list, t_list)
plt.xlabel("Probability of Infection")
plt.ylabel("Length of Infection (time steps)")
plt.savefig("soundcould_infection_length.png")
plt.clear()
plt.title("Infected Percentage vs Probability of Infection")
plt.plot(p_list, infected_list)
plt.xlabel("Probability of Infection")
plt.ylabel("Percent Infected")
plt.savefig("soundcould_infection_percent.png")
plt.clear()
def plot_multiple_figure(figure_names: vector,
plot_handler,
tight_layout=False,
saved_path=None,
max_cols=-1):
# plt.gca()
N = figure_names.length
if N == 0:
return False
if max_cols == -1:
cols = math.floor(math.sqrt(N))
else:
cols = min(math.floor(math.sqrt(N)), max_cols)
rows = (N + cols - 1) // cols
plt.clear()
fig = plt.figure(figsize=(cols * 4, rows * 4))
for index in range(N):
ax = plt.subplot(rows, cols, index + 1)
plot_handler[figure_names[index]](ax)
if tight_layout:
plt.tight_layout()
if saved_path is not None:
if saved_path.endswith("pdf"):
with PdfPages(saved_path, "w") as f:
plt.savefig(f, format="pdf")
else:
plt.savefig(saved_path, dpi=300)
else:
plt.show()
return True
def plotTSP(tsp : TSP, plotArea):
plt = plotArea.figure.gca()
plt.clear()
xList = []
yList = []
for city in tsp.cityList:
xList.append(city.posX)
yList.append(city.posY)
plt.plot(xList, yList)
plotArea.Layout()
def _plot_x_y(self, ix, iy, xlabel, ylabel, scatter, modes=None,
fig=None, axes=None,
xlim=None, ylim=None,
show=True, clear=False, legend=True,
png_filename=None,
**kwargs):
"""builds the plot"""
self.fix()
if kwargs is None:
kwargs = {}
modes, imodes = _get_modes_imodes(self.modes, modes)
if fig is None:
fig = plt.figure()
axes = fig.add_subplot(111)
symbols = self.symbols
for i, imode, mode in zip(count(), imodes, modes):
symbol = symbols[i]
freq = self.results[imode, :, self.ifreq].ravel()
xs = self.results[imode, :, ix].ravel()
ys = self.results[imode, :, iy].ravel()
iplot = np.where(freq != np.nan)
#iplot = np.where(freq > 0.0)
axes.plot(xs[iplot], ys[iplot], symbol, label='Mode %i' % mode, markersize=0)
if scatter:
scatteri = np.linspace(.75, 50., len(xs))
#assert symbol[2] == '-', symbol
axes.scatter(xs[iplot], ys[iplot], s=scatteri, color=symbol[0], marker=symbol[1])
axes.grid(True)
axes.set_xlabel(xlabel)
axes.set_ylabel(ylabel)
if xlim:
axes.set_xlim(xlim)
if ylim:
axes.set_ylim(ylim)
title = 'Subcase %i' % self.subcase
if png_filename:
title += '\n%s' % png_filename
fig.suptitle(title)
if legend:
axes.legend(**kwargs)
if show:
plt.show()
if png_filename:
plt.savefig(png_filename)
if clear:
plt.clear()
return axes
def plot_transient(sample_df, sample_name, save_dir):
# plot the time sequence parameter
plt.figure(figsize=(16, 12))
plt.subplot(221)
plt.plot(sample_df.index.values, sample_df.gateSignalVoltage.values)
plt.title('Gate Signal Voltage', fontsize='xx-large')
plt.xlabel('Time /10ns', fontsize='xx-large')
plt.ylabel('Voltage /V', fontsize='xx-large')
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.ylim([-2, 12])
plt.tight_layout()
# plt.savefig( 'transient_picture/gateSignalVoltage'+sample_name+'.png')
plt.subplot(222)
plt.plot(sample_df.index.values, sample_df.gateEmitterVoltage.values)
plt.title('Gate Emitter Voltage', fontsize='xx-large')
plt.xlabel('Time /10ns', fontsize='xx-large')
plt.ylabel('Voltage /V', fontsize='xx-large')
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.ylim([-2, 12])
plt.tight_layout()
# plt.savefig( 'transient_picture/gateEmitterVoltage'+sample_name+'.png')
# plot the time sequence parameter
plt.subplot(223)
plt.plot(sample_df.index.values, sample_df.collectorEmitterVoltage.values)
plt.title('Collector Emitter Voltage', fontsize='xx-large')
plt.xlabel('Time /10ns', fontsize='xx-large')
plt.ylabel('Voltage /V', fontsize='xx-large')
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.ylim([-2, 12])
plt.tight_layout()
# plt.savefig('transient_picture//collectorEmitterVoltage'+sample_name+'.png')
# collectorEmitterCurrentSingal
# plot the time sequence parameter
plt.subplot(224)
plt.plot(sample_df.index.values,
sample_df.collectorEmitterCurrentSingal.values)
plt.title('Collector Emitter Current', fontsize='xx-large')
plt.xlabel('Time /10ns', fontsize='xx-large')
plt.ylabel('Current /A', fontsize='xx-large')
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.ylim([-2, 12])
plt.tight_layout()
plt.savefig(save_dir + sample_name + '.png')
plt.clear("all")
def _plot_x_y(self, ix, iy, xlabel, ylabel, scatter, modes=None,
fig=None,
xlim=None, ylim=None,
show=True, clear=False, legend=True,
png_filename=None,
**kwargs):
"""builds the plot"""
self.fix()
if kwargs is None:
kwargs = {}
modes, imodes = self._get_modes_imodes(modes)
if fig is None:
fig = plt.figure()
axes = fig.add_subplot(111)
self._set_xy_limits(xlim, ylim)
symbols = self.symbols
for i, imode, mode in zip(count(), imodes, modes):
symbol = symbols[i]
freq = self.results[imode, :, self.ifreq].ravel()
real = self.results[imode, :, ix].ravel()
imag = self.results[imode, :, iy].ravel()
iplot = np.where(freq != np.nan)
#iplot = np.where(freq > 0.0)
axes.plot(real[iplot], imag[iplot], symbol, label='Mode %i' % mode, markersize=0)
if scatter:
s = np.linspace(.75, 50., len(real))
#assert symbol[2] == '-', symbol
axes.scatter(real[iplot], imag[iplot], s=s, color=symbol[0], marker=symbol[1])
axes.grid(True)
axes.set_xlabel('Eigenvalue (Real)')
axes.set_ylabel('Eigenvalue (Imaginary)')
title = 'Subcase %i' % self.subcase
if png_filename:
title += '\n%s' % png_filename
fig.suptitle(title)
if legend:
axes.legend(**kwargs)
if show:
plt.show()
if png_filename:
plt.savefig(png_filename)
if clear:
plt.clear()
def animate(i):
pullData = open("sampleText.txt", "r").read()
dataList = pullData.split('\n')
xList = []
yList = []
for eachLine in dataList:
if len(eachLine) > 1:
x, y = eachLine.split(',')
x = datetime.strptime(x, "%Y-%m-%d %H:%M:%S.%f")
x = matplotlib.dates.date2num(x)
xList.append(int(x))
yList.append(int(y))
plt.clear()
plt.plot(xList, yList)
plt.show()
def update(i):
var_t_val=var_t_vals[i]
label = str(var_t_val)
# print(label)
# Update the line and the axes (with a new xlabel). Return a tuple of
# "artists" that have to be redrawn for this frame.
plt.clear()
var_x_vals = (merge_df.loc[merge_df[var_t]==var_t_val,var_x]).reset_index(drop=True)
var_y_vals = (merge_df.loc[merge_df[var_t]==var_t_val,(var_y+"_"+source)]).reset_index(drop=True)
# line.set_xdata(var_x_vals)
# line.set_ydata(var_y_vals)
plt.plot(var_x_vals, var_y_vals, color=colour_models(var_y))
#line, = ax.plot(var_x_vals, var_y_vals, 'ro', linewidth=2)
plt.set_xlabel(label)
#plt.title(label)
return line, ax
def draw_plot(x,
y,
title="plot",
xlabel="x axis",
ylabel="y axis",
legends=[]):
coef_margin = 0.1
x_len = np.max(x) - np.min(x)
y_len = np.max(y) - np.min(y)
plt.clear()
plt.title(title)
plt.axis([np.min(x) - coef_margin * x_len, np.max(x) + coef_margin * x_len, \
np.min(y) - coef_margin * y_len, np.max(y) + coef_margin * y_len])
plt.xlabel(xlabel)
plt.ylabel(ylabel)
plt.plot(x, y)
plt.show()
def plot_graph(self):
self.ui.tempGraph.plot(self.idx_list, self.temp_list)
plt = self.ui.tempGraph.canvas.ax
plt.clear()
plt.set_xlabel('Latest 10 values')
if self.mode == "C":
plt.set_ylabel('Temperature (Celsius)')
plt.set_title('Temperature') #| Average = {0:0.1f} deg C
elif self.mode == "F":
plt.set_ylabel('Temperature (Fahrenheit)')
plt.set_title('Temperature') # | Average = {0:0.1f} deg F.format(self.avgT).format(self.avgT)
self.ui.tempGraph.canvas.draw()
self.ui.humGraph.plot(self.idx_list, self.hum_list)
plt2 = self.ui.humGraph.canvas.ax
plt2.clear()
plt2.set_xlabel('Latest 10 values')
plt2.set_ylabel('Humidity (%)')
plt2.set_title('Humidity') #| Average = {0:0.1f} %.format(self.avgH)
self.ui.humGraph.canvas.draw()
def plot_vg_vf(self, fig=None, modes=None, show=None, png_filename=None,
clear=False, legend=None,
xlim=None, ylim_damping=None, ylim_freq=None):
"""
Make a V-g and V-f plot
Parameters
----------
modes : List[int] / int ndarray; (default=None -> all)
the modes; typically 1 to N
"""
self.fix()
if fig is None:
fig = plt.figure() # figsize=(12,9), self.subcase
gridspeci = gridspec.GridSpec(2, 4)
damp_axes = fig.add_subplot(gridspeci[0, :3])
freq_axes = fig.add_subplot(gridspeci[1, :3], sharex=damp_axes)
#self._set_xy_limits(xlim, ylim)
modes, imodes = self._get_modes_imodes(modes)
symbols = self.symbols
_kvelocity, velocity_units = self._get_unit_factor('velocity')
legend_items = ['Mode %i' % mode for mode in modes]
for i, imode, mode in zip(count(), imodes, modes):
vel = self.results[imode, :, self.ivelocity].ravel()
damping = self.results[imode, :, self.idamping].ravel()
freq = self.results[imode, :, self.ifreq].ravel()
#iplot = np.where(freq > 0.0)
#damp_axes.plot(vel, damping, symbols[i], label='Mode %i' % mode)
#freq_axes.plot(vel, freq, symbols[i])
iplot = np.where(freq != np.nan)
damp_axes.plot(vel[iplot], damping[iplot], symbols[i], label='Mode %i' % mode)
freq_axes.plot(vel[iplot], freq[iplot], symbols[i])
damp_axes.set_xlabel('Velocity [%s]' % velocity_units)
damp_axes.set_ylabel('Damping')
damp_axes.grid(True)
if xlim is not None:
damp_axes.set_xlim(xlim)
if ylim_damping is not None:
damp_axes.set_ylim(ylim_damping)
freq_axes.set_xlabel('Velocity [%s]' % velocity_units)
freq_axes.set_ylabel('Frequency [Hz]')
freq_axes.grid(True)
if xlim is not None:
freq_axes.set_xlim(xlim)
if ylim_freq is not None:
freq_axes.set_ylim(ylim_freq)
title = 'Subcase %i' % self.subcase
if png_filename:
title += '\n%s' % png_filename
damp_axes.set_title(title)
#plt.suptitle(title)
if legend:
damp_axes.legend(legend_items, fontsize=10, bbox_to_anchor=(1.125, 1.), loc=2)
#fig.subplots_adjust(hspace=0.25)
#fig.subplots_adjust(hspace=.5)
#plt.legend()
#damp_axes.legend(legend_items, bbox_to_anchor=anchor, ncol=2)
#fig.subplots_adjust(hspace=0.25)
#fig.subplots_adjust(hspace=.5)
if show:
plt.show()
if png_filename:
plt.savefig(png_filename)
if clear:
plt.clear()
def update():
area_from_ξ(ξ, show=True, params=p)
plt.clear('all')
def destroy_fig(self):
plt.clear(self.fig)
def inline_train(self, input, output):
'''
Function to execute the inline training of the RHONN
Parameters
----------
input - an iterable with the input given to the system.
output - an iterable with the output measured from the system.
'''
# Initial values
if not self.inline_started:
self.s_x = np.zeros(self.in_size)
self.weights = np.random.rand(self.out_size, self.num_z)
self.error = np.zeros(self.out_size)
self.z = np.random.rand(self.out_size, self.num_z)
self.x_estimated = np.zeros(self.out_size)
self.k = 0
# Var to save the training history
self.inputs = []
self.outputs = []
self.net_outputs = []
self.errors = []
# Create figures for ploting data
fig_error = plt.figure('Error')
error_ax = fig_error.add_subplot(111)
figs = [
plt.figure('State variable {}'.format(n + 1))
for n in range(self.out_size)
]
axes = [figs[n].add_subplot(111) for x in range(self.out_size)]
self.inline_started = True
self.inputs.append(input)
self.outputs.append(output)
# Applying activation functions
for n in range(len(self.activation_funcs)):
self.s_x[n] = self.activation_funcs[n](input[n])
# TODO: Fix the elements in the Z vector
# Calculating network output
if self.z_structure is not None:
self.z = self.calculate_z(self.s_x)
else:
self.z = np.array([self.s_x for x in range(self.num_z)])
self.x_estimated = [
np.dot(self.weights[n].T, self.z[n]) for n in range(self.out_size)
]
# Calculating the error
self.error = np.subtract(output, self.x_estimated)
# Save results
self.net_outputs.append(self.x_estimated)
self.errors.append(self.error)
if self.debug:
print self.str_result.format(k, output, self.x_estimated,
self.error)
# Calculating the adjustment for all the array with EFK
if self.fnc_weights_adjsmnt is None:
weights_res = self.EKF_weights_adjustment(self.z, self.weights,
self.error)
estimated_weights = weights_res[0]
self.EKF_P = weights_res[1]
else:
estimated_weights = self.fnc_weights_adjsmnt(
self.weights, self.error)
# Save the new weights
self.weights = estimated_weights
# Save the time units elapsed
self.k += 1
# Ploting data
if self.plot_data:
plt.clear()
for n in range(self.out_size):
line = error_ax.plot(self.error[:, n],
label='State var {}'.format(n + 1))
plt.legend()
for n in range(self.out_size):
axes[n].clear()
line_1, = axes[n].plot(self.outputs[:, [n]], label='System')
line_2, = axes[n].plot(self.net_out[:, [n]],
linestyle='--',
label='RHONN',
color='red')
plt.legend()
plt.show()
pass
print(INSTR_IDX)
if INSTR_IDX == 1 and button == 'next':
frame.create_image(0, 0, image=img, anchor=NW)
root.title("Instructions- page 2/2")
INSTR_IDX += 1
elif INSTR_IDX == 2 and button == 'prev':
frame.create_image(0, 0, image=img, anchor=NW)
root.title("Instructions- page 1/2")
INSTR_IDX -= 1
elif INSTR_IDX == 1 and button == 'prev':
return
else:
root.destroy()
"""--------------------------------------------------
this function creates buttons 'next','reset','hint'
in GUI, after user clears the current choice (after
plt.clear() is called)
...
no params for this function
--------------------------------------------------"""
def make_buttons():
global im_hint
global im_hint_off_1
global show_suggestions_axes
next_button_axes = plt.axes([0.91, 0.46, 0.08, 0.08])
im_next = pilim.open("imgs/Toggle_next.png")
next_graph_but = matButton(next_button_axes,
def plot_vg_vf(self,
fig=None,
damp_axes=None,
freq_axes=None,
modes=None,
show=None,
plot_type='tas',
png_filename=None,
clear=False,
legend=None,
xlim=None,
ylim_damping=None,
ylim_freq=None,
nopoints=False):
"""
Make a V-g and V-f plot
Parameters
----------
modes : List[int] / int ndarray; (default=None -> all)
the modes; typically 1 to N
"""
#self.fix()
if fig is None:
fig = plt.figure() # figsize=(12,9), self.subcase
gridspeci = gridspec.GridSpec(2, 4)
damp_axes = fig.add_subplot(gridspeci[0, :3])
freq_axes = fig.add_subplot(gridspeci[1, :3], sharex=damp_axes)
#self._set_xy_limits(xlim, ylim)
modes, imodes = _get_modes_imodes(self.modes, modes)
symbols = self.symbols
legend_items = ['Mode %i' % mode for mode in modes]
ix, xlabel = self._plot_type_to_ix_xlabel(plot_type)
for i, imode, mode in zip(count(), imodes, modes):
vel = self.results[imode, :, ix].ravel()
damping = self.results[imode, :, self.idamping].ravel()
freq = self.results[imode, :, self.ifreq].ravel()
#iplot = np.where(freq > 0.0)
#damp_axes.plot(vel, damping, symbols[i], label='Mode %i' % mode)
#freq_axes.plot(vel, freq, symbols[i])
#iplot = np.where(freq != np.nan)
#damp_axes.plot(vel[iplot], damping[iplot], symbols[i], label='Mode %i' % mode)
#freq_axes.plot(vel[iplot], freq[iplot], symbols[i])
if symbols and not nopoints:
symbol = symbols[i]
damp_axes.plot(vel, damping, symbol, label='Mode %i' % mode)
freq_axes.plot(vel, freq, symbol)
else:
damp_axes.plot(vel, damping, label='Mode %i' % mode)
freq_axes.plot(vel, freq)
damp_axes.set_xlabel(xlabel)
freq_axes.set_xlabel(xlabel)
damp_axes.set_ylabel('Damping')
damp_axes.grid(True)
if xlim is not None:
damp_axes.set_xlim(xlim)
if ylim_damping is not None:
damp_axes.set_ylim(ylim_damping)
freq_axes.set_ylabel('Frequency [Hz]')
freq_axes.grid(True)
if xlim is not None:
freq_axes.set_xlim(xlim)
if ylim_freq is not None:
freq_axes.set_ylim(ylim_freq)
title = 'Subcase %i' % self.subcase
if png_filename:
title += '\n%s' % png_filename
damp_axes.set_title(title)
#plt.suptitle(title)
if legend:
damp_axes.legend(legend_items,
fontsize=10,
bbox_to_anchor=(1.125, 1.),
loc=2)
#fig.subplots_adjust(hspace=0.25)
#fig.subplots_adjust(hspace=.5)
#plt.legend()
#damp_axes.legend(legend_items, bbox_to_anchor=anchor, ncol=2)
#fig.subplots_adjust(hspace=0.25)
#fig.subplots_adjust(hspace=.5)
if show:
plt.show()
if png_filename:
plt.savefig(png_filename)
if clear:
plt.clear()
# Thu, 08 Jun 2017 01:04:08
import numpy as np
import matplotlib.pyplot as plt
import time
plt.axis([0, 10, 0, 1])
plt.ion()
for i in range(10):
y = np.random.random()
plt.scatter(i, y)
plt.pause(0.01)
time.sleep(1)
# Thu, 08 Jun 2017 01:04:27
plt.clear()
# Thu, 08 Jun 2017 01:04:30
plt.cla()
# Thu, 08 Jun 2017 01:04:38
import numpy as np
import matplotlib.pyplot as plt
import time
plt.axis([0, 10, 0, 1])
plt.ion()
for i in range(10):
y = np.random.random()
plt.scatter(i, y)
plt.pause(0.01)
time.sleep(1)
def listofpeakinfo(x,y,indexes,samplename):#x and y are np.arrays
peakdata=[]
try:
plt.clear()
except:
pass
plt.figure(figsize=(10,6))
plt.plot(x,y,'black',label=samplename)
for item in range(len(indexes)):
nbofpoints=80#on each side of max position
while(1):
try:
x0=x[indexes[item]-nbofpoints:indexes[item]+nbofpoints]
y0=y[indexes[item]-nbofpoints:indexes[item]+nbofpoints]
#baseline height
bhleft=np.mean(y0[:20])
bhright=np.mean(y0[-20:])
baselineheightatmaxpeak=(bhleft+bhright)/2
if abs(bhleft-bhright)<50:#arbitrary choice of criteria...
#find FWHM
d=y0-((max(y0)-bhright)/2)
ind=np.where(d>bhright)[0]
hl=(x0[ind[0]-1]*y0[ind[0]]-y0[ind[0]-1]*x0[ind[0]])/(x0[ind[0]-1]-x0[ind[0]])
ml=(y0[ind[0]-1]-hl)/x0[ind[0]-1]
yfwhm=((max(y0)-baselineheightatmaxpeak)/2)+baselineheightatmaxpeak
xleftfwhm=(yfwhm - hl)/ml
hr=(x0[ind[-1]]*y0[ind[-1]+1]-y0[ind[-1]]*x0[ind[-1]+1])/(x0[ind[-1]]-x0[ind[-1]+1])
mr=(y0[ind[-1]]-hr)/x0[ind[-1]]
xrightfwhm=(yfwhm - hr)/mr
FWHM=abs(xrightfwhm-xleftfwhm)
Peakheight=max(y0)-baselineheightatmaxpeak
center=x[indexes[item]]
plt.plot(x0, y0, 'red')
plt.plot([x0[0],x0[-1]],[bhleft,bhright],'blue')
# plt.plot(x0,y0,ms=0)
plt.plot([xleftfwhm,xrightfwhm],[yfwhm,yfwhm],'green')
plt.text(center,max(y0)+200,str('%.1f' % float(center)),rotation=90,verticalalignment='bottom',horizontalalignment='center',multialignment='center')
peakdata.append([center,FWHM,Peakheight])
break
else:
if nbofpoints>=15:
nbofpoints-=10
else:
print("indexerror unsolvable")
break
except IndexError:
if nbofpoints>=15:
nbofpoints-=10
else:
print("indexerror unsolvable")
break
plt.scatter(x[indexes],y[indexes],c='red',s=12)
plt.legend()
plt.ylabel("Intensity (a.u.)")
plt.xlabel("2\u0398 (degree)")
plt.savefig(samplename+'.pdf')
# plt.show()
plt.close()
return peakdata
def plotGraph(xList, yList, name, plotArea):
plt = plotArea.figure.gca()
plt.clear()
plotArea.figure.axes.clear()
plt.plot(xList, yList)
plotArea.Layout()
def plot_option_scores():
xs = [10 + 10 * x for x in range(1000)]
scores = bruteforce_option_stats()
top_scorers = {}
worst_scorers = {}
score_history = {}
# for option_ids, option_scores in list(scores.items()):
# scores[option_ids] = np.log(1 + option_scores)
for option_ids, option_scores in scores.items():
for score_idx, nr_iter in enumerate(xs):
option_score = option_scores[score_idx]
top_scorers[nr_iter] = (option_ids, option_score)
worst_scorers[nr_iter] = (option_ids, option_score)
score_history[nr_iter] = []
break
for option_ids, option_scores in tqdm.tqdm(scores.items()):
for score_idx, nr_iter in enumerate(xs):
option_score = option_scores[score_idx]
if top_scorers[nr_iter][1] < option_score:
top_scorers[nr_iter] = (option_ids, option_score)
if worst_scorers[nr_iter][1] > option_score:
worst_scorers[nr_iter] = (option_ids, option_score)
score_history[nr_iter].append(option_score)
scores = bruteforce_options()
no_options = scores[(None, None, None, None)]
xs = [10, 30, 60, 200, 500, 1000, 2000, 4000, 8000, xs[-1]]
import seaborn as sns
sns.set(color_codes=True)
data = []
# xs = sorted(score_history.keys())
for x in xs:
data.append(score_history[x])
plt.figure(1, figsize=(15, 5))
y_low, y_high = worst_scorers[xs[0]][1], top_scorers[xs[-1]][1]
y_high = np.log(1000)
plt.ylim(ymin=y_low, ymax=y_high)
# plt.subplot(
data = np.array(data)
no_options_show = []
xs_full = [10 + 10 * x for x in range(1000)]
for idx, x in enumerate(xs_full):
if x in xs:
no_options_show.append(no_options[idx])
percentiles_ranges = [1, 50, 90, 95, 99, 99.9, 100]
percentiles = [{} for nr_iter in percentiles_ranges]
for nr_iter in tqdm.tqdm(xs, desc="percentiels"):
for idx, perc in enumerate(percentiles_ranges):
percentiles[idx][nr_iter] = np.percentile(score_history[nr_iter],
perc)
x_labels = [str(idx) + "_" + str(x) + "_iter" for idx, x in enumerate(xs)]
# plt.plot(x_labels, data)
plt.plot(x_labels, data.mean(axis=1), 'o', label="mean")
# print(percentiles)
for idx, perc in tqdm.tqdm(enumerate(percentiles_ranges),
desc="plot percs"):
ys = [percentiles[idx][x] for x in xs]
plt.plot(x_labels, ys, 'o', label="perc:" + str(perc))
plt.legend(loc='upper right')
plt.title("percentiles")
plt.savefig("percentiles.png")
plt.plot(x_labels, no_options_show, label="no options")
plt.legend(loc='upper left')
plt.title("percentiles and no options")
plt.savefig("percentiles_and_no_options.png")
plt.clear()
x_labels = [
str(idx) + "_" + str(x) + "%"
for idx, x in enumerate(percentiles_ranges)
]
for nr_iter in tqdm.tqdm(percentiles[0].keys(), desc="perc vs score"):
ys = [
percentiles[percentiles_ranges.index(perc)][nr_iter]
for perc in percentiles_ranges
]
plt.plot(x_labels, ys, 'o', label="iter:" + str(nr_iter))
plt.legend(loc='upper left')
plt.title("perc vs score")
plt.savefig("perc_vs_score.png")
plt.clear()
x_labels = [str(idx) + "_" + str(x) + "_iter" for idx, x in enumerate(xs)]
ys = [
percentiles[percentiles_ranges.index(100)][nr_iter] for nr_iter in xs
]
plt.plot(x_labels, ys, '-', label="best")
plt.legend(loc='upper left')
plt.title("best")
plt.savefig("best.png")
# plt.clear()
plt.plot(x_labels, no_options_show, label="no options")
plt.legend(loc='upper left')
plt.title("best and no options")
plt.savefig("best_and_no_options.png")
plt.clear()
# ys = [percentiles[percentiles_ranges.index(1)][nr_iter] for nr_iter in xs]
# plt.plot(x_labels, ys, label="perc:1%")
# ys = [percentiles[percentiles_ranges.index(90)][nr_iter] for nr_iter in xs]
# plt.plot(x_labels, ys, label="perc:90%")
ys = [percentiles[percentiles_ranges.index(50)][nr_iter] for nr_iter in xs]
plt.plot(x_labels, ys, label="perc:50%")
plt.legend(loc='upper left')
plt.title("perc:50%")
plt.savefig("perc:50%.png")
plt.clear()
# ys = [percentiles[percentiles_ranges.index(95)][nr_iter] for nr_iter in xs]
# plt.plot(x_labels, ys, label="perc:95%")
ys = [percentiles[percentiles_ranges.index(99)][nr_iter] for nr_iter in xs]
plt.plot(x_labels, ys, label="perc:99%")
plt.legend(loc='upper left')
plt.title("perc:99%")
plt.savefig("perc:99%.png")
plt.clear()
# plt.legend(loc='upper left')
# plt.show()
cutoff = {
nr_iter: percentiles[percentiles_ranges.index(99.9)][nr_iter]
for nr_iter in xs
}
best_sets = {}
for nr_iter in xs:
best_sets[nr_iter] = []
print("best sets")
for option_ids, option_scores in tqdm.tqdm(scores.items(),
desc="best sets"):
for iter_idx, option_score in enumerate(option_scores):
nr_iter = xs_full[iter_idx]
if nr_iter in xs:
if option_score > cutoff[nr_iter]:
best_sets[nr_iter].append((option_ids, option_score))
for nr_iter in best_sets.copy().keys():
best_sets[nr_iter].sort(key=lambda _x: -_x[1])
import pprint
pprint.pprint(best_sets)
import pickle
with open("best_sets.txt", "wb") as fout:
pickle.dump(pprint.pformat(best_sets), fout)
def plot_single_run_over_time(single_run_data, directory):
plt.clear()
plt.plot(single_run_data["Generation"],
np.log(single_run_data["Average_Fitness"]))
plt.savefig(directory + "/single_run_over_time.png")
import matplotlib.colors as colors
import math
print 'Script started'
fig = plt.figure()
ax = fig.add_subplot(111)
ratio = 1.0 / 3.0
count = math.ceil(math.sqrt(len(colors.cnames)))
x_count = count * ratio
y_count = count / ratio
x = 0
y = 0
w = 1 / x_count
h = 1 / y_count
print '\tStarting loop...'
for c in colors.cnames:
pos = (x / x_count, y / y_count)
ax.add_patch(patches.Rectangle(pos, w, h, color=c))
ax.annotate(c, xy=pos)
if y >= y_count - 1:
x += 1
y = 0
else:
y += 1
print '\tloop ended'
plt.show()
plt.clear('all')
print 'Script finished successfully'
def _plot_x_y2(self,
ix,
iy1,
iy2,
xlabel,
ylabel1,
ylabel2,
scatter,
modes=None,
fig=None,
axes1=None,
axes2=None,
xlim=None,
ylim1=None,
ylim2=None,
show=True,
clear=False,
legend=True,
png_filename=None,
**kwargs):
"""builds the plot"""
self.fix()
if kwargs is None:
kwargs = {}
modes, imodes = _get_modes_imodes(self.modes, modes)
if fig is None:
fig = plt.figure()
gridspeci = gridspec.GridSpec(2, 4)
axes1 = fig.add_subplot(gridspeci[0, :3])
axes2 = fig.add_subplot(gridspeci[1, :3], sharex=axes1)
if xlim:
axes1.set_xlim(xlim)
if ylim1:
axes1.set_ylim(ylim1)
if ylim2:
axes2.set_ylim(ylim2)
symbols = self.symbols
for i, imode, mode in zip(count(), imodes, modes):
symbol = symbols[i]
freq = self.results[imode, :, self.ifreq].ravel()
xs = self.results[imode, :, ix].ravel()
y1s = self.results[imode, :, iy1].ravel()
y2s = self.results[imode, :, iy2].ravel()
iplot = np.where(freq != np.nan)
#iplot = np.where(freq > 0.0)
axes1.plot(xs[iplot],
y1s[iplot],
symbol,
label='Mode %i' % mode,
markersize=0)
axes2.plot(xs[iplot],
y2s[iplot],
symbol,
label='Mode %i' % mode,
markersize=0)
if scatter:
scatteri = np.linspace(.75, 50., len(xs))
#assert symbol[2] == '-', symbol
axes1.scatter(xs[iplot],
y1s[iplot],
s=scatteri,
color=symbol[0],
marker=symbol[1])
axes2.scatter(xs[iplot],
y2s[iplot],
s=scatteri,
color=symbol[0],
marker=symbol[1])
axes1.grid(True)
axes1.set_xlabel(xlabel)
axes1.set_ylabel(ylabel1)
axes2.grid(True)
axes2.set_xlabel(xlabel)
axes2.set_ylabel(ylabel2)
title = 'Subcase %i' % self.subcase
if png_filename:
title += '\n%s' % png_filename
fig.suptitle(title)
if legend:
axes1.legend(**kwargs)
if show:
plt.show()
if png_filename:
plt.savefig(png_filename)
if clear:
plt.clear()
def plot_single_run_over_time(single_run_data, directory):
plt.clear()
plt.plot(single_run_data["Generation"], np.log(single_run_data["Average_Fitness"]))
plt.savefig(directory+"/single_run_over_time.png")
plt.minorticks_on()
plt.grid(b=True, which='minor', alpha=0.2)
#interpolations for imshow of images
# https://matplotlib.org/gallery/images_contours_and_fields/interpolation_methods.html
plt.imshow(image, interpolation='nearest')
#marker type
#https://matplotlib.org/3.1.0/api/markers_api.html
# 's' = square
# '.' = dot
# 'x' = cross
### CLEARING CHARTS---------------------------------------
plt.cla() #clears an axis, i.e. the currently active axis in the current figure. It leaves the other axes untouched.
plt.clear() #clears the entire current figure with all its axes, but leaves the window opened, such that it may be reused for other plots.
plt.close() #closes a window, which will be the current window, if not specified otherwise.
hline = plt.axhline(y = line, color = "red", lw=1)
hline.remove() #remove an object in the chart
# remove x & y axes
plt.axis('off')
### SAVE CHART IMAGE---------------------------------------
plt.plot(range(10))
plt.savefig('testplot.png', dpi=300) #save as png
import Image
Image.open('testplot.png').save('testplot.jgp', 'JPEG') #save as jpeg using Pillow
def hist_plot_ns(chrom_len_dict, ns_window_dict, window_size, scaffold_list):
"""TODO: Docstring for hist_plot_ns.
:chom_len_dict: TODO
:ns_window_dict: TODO
:window_size: TODO
:returns: TODO
"""
if scaffold_list == None:
for chrom, val in ns_window_dict.items():
get_chrom_len = int(chrom_len_dict[chrom])
create_x_range = range(0, get_chrom_len, window_size)
#Enumerate Bins correctly
x_pos = [i for i, _ in enumerate(create_x_range)]
print("Plotting %s" % chrom)
#print(len(create_x_range))
#print(len(val))
#print(x_pos)
#Titles and width
x_label = "Window size of %s" % str(window_size)
y_label = "N freq, max with window size (%s)" % str(window_size)
title = "Distribution of Ns in %s with length %s bp" % (str(chrom), \
str(get_chrom_len))
bar_width = 0.5
#Create and show plot
plt.bar(x_pos, val, bar_width)
plt.xlabel(x_label)
plt.ylabel(y_label)
plt.title(title)
#plt.xticks(rotation=90)
#plt.xticks(np.arange(min(x_pos), max(x_pos)+1, 1))
creat_file_name = chrom + '_plot.pdf'
plt.savefig(creat_file_name)
plt.gcf().clear()
#plt.show()
#If given a list to work with, only pring and write the given ones
elif scaffold_list != None:
for chrom, val in ns_window_dict.items():
if chrom in scaffold_list:
get_chrom_len = int(chrom_len_dict[chrom])
create_x_range = range(0, get_chrom_len, window_size)
#Enumerate Bins correctly
x_pos = [i for i, _ in enumerate(create_x_range)]
print("Plotting %s" % chrom)
#print(len(create_x_range))
#print(len(val))
#print(x_pos)
#Titles and width
x_label = "Window size of %s" % str(window_size)
y_label = "N freq, max with window size (%s)" % str(
window_size)
title = "Distribution of Ns in %s with length %s bp" % (str(chrom), \
str(get_chrom_len))
bar_width = 0.5
#Create and show plot
plt.bar(x_pos, val, bar_width)
plt.xlabel(x_label)
plt.ylabel(y_label)
plt.title(title)
#plt.xticks(rotation=90)
#plt.xticks(np.arange(min(x_pos), max(x_pos)+1, 1))
creat_file_name = chrom + '_plot.pdf'
plt.savefig(creat_file_name)
#Clear the plot
plt.clear()
else:
pass
