def emfTitration(ax, massAcid, emf, massSample, concAcid, alk_emf0, alkGuess,
rgb, sublabel):
"""EMF change as acid is added throughout a titration."""
ax.axvline(1e3 * alk_emf0['x'][0] * massSample / concAcid,
color=_rgb_final,
linestyle='--',
zorder=1)
ax.axvline(1e3 * alkGuess * massSample / concAcid,
color=_rgb_guess,
linestyle='--',
zorder=1)
ax.scatter(massAcid * 1e3,
emf,
c=rgb,
edgecolors='k',
clip_on=False,
zorder=2)
ax.set_xlim([0, np_max(massAcid) * 1e3])
yrange = np_max(emf) - np_min(emf)
ax.set_ylim([np_min(emf) - yrange * 0.05, np_max(emf) + yrange * 0.05])
ax.set_xlabel('Acid mass / g')
ax.set_ylabel('EMF / mV')
ax.set_title('{} Final EMF$^\circ$ = {:.2f} mV'.format(
sublabel, alk_emf0['x'][1]),
fontsize=10)
return ax
def update(self, abs_tol=1e-5, rel_tol=1e-3):
'''
during the update step, you calculate the gradient of Q
and update w and b.
'''
# accumulate the decay rates, in order to correct the averages
self.beta_m_ac *= self.beta_m_ac
#The update should run after
#FFANN.feedForward() and FFANN.backPropagation().
#these will be used to determine if the stopping conditions are satisfied
_w2 = 0
_check = 0
self.Q.randomDataPoint()
for l in range(self.Q.model.total_layers - 1):
for j in range(self.Q.model.nodes[l + 1]):
for i in range(self.Q.model.nodes[l]):
#get the grad of the loss. The results should be stored in loss.dQdw and loss.dQdb
#This way it should be easy to update the weights and biases of FFANN
self.Q.grad(l, j, i)
self.mw[l][j][i] = self.beta_m * self.mw[l][j][i] + (
1 - self.beta_m) * self.Q.dQdw
self.vw_max[l][j][i] = np_max([
self.beta_v * self.vw_max[l][j][i],
np_abs(self.Q.dQdw)
])
dw = self.alpha / (self.vw_max[l][j][i] +
self.epsilon) * self.mw[l][j][i] / (
1 - self.beta_m_ac)
#update the weight using loss.dQdw
self.Q.model.addToWeight(l, j, i, -dw)
_w2 = abs_tol + np_abs(
self.Q.model.weights[l][j][i]) * rel_tol
_check += (dw / _w2) * (dw / _w2)
#update the bias using loss.dQdb (it is the same for all i, so don't run loss.grad again).
self.mb[l][j] = self.beta_m * self.mb[l][j] + (
1 - self.beta_m) * self.Q.dQdb
self.vb_max[l][j] = np_max(
[self.beta_v * self.vb_max[l][j],
np_abs(self.Q.dQdb)])
dw = self.alpha / (self.vb_max[l][j] + self.epsilon
) * self.mb[l][j] / (1 - self.beta_m_ac)
self.Q.model.addToBias(l, j, -dw)
_w2 = abs_tol + np_abs(self.Q.model.biases[l][j]) * rel_tol
_check += (dw / _w2) * (dw / _w2)
_check = np_sqrt(1. / self.Q.N * _check)
return _check
def plot_optimized_network(network, blocking=True, save_plot=True):
_title = 'Optimized network vs. non-optimized'
figure(_title)
graph = network.network_plot
plot_layout = spring_layout(graph)
balls = network.get_ball_distribution()
print(network.ref_distribution)
print(balls)
min_val = 0
max_val = max([np_max(network.ref_distribution), np_max(balls)])
cmap = cm.Greys
color_vals = cm.ScalarMappable(cmap=cmap,
norm=Normalize(vmin=min_val, vmax=max_val))
color_vals._A = []
subplot(2, 1, 1)
title('Initial network')
colorbar(color_vals)
draw_networkx_edges(graph, plot_layout, alpha=.3)
draw_networkx_nodes(graph,
plot_layout,
node_size=100,
edgecolors='k',
node_color=network.ref_distribution,
cmap=cmap,
vmin=min_val,
vmax=max_val)
axis('off') # Disable axis
subplot(2, 1, 2)
title('Optimized network')
colorbar(color_vals)
draw_networkx_edges(graph, plot_layout, alpha=.3)
draw_networkx_nodes(graph,
plot_layout,
node_size=100,
edgecolors='k',
node_color=balls,
cmap=cmap,
vmin=min_val,
vmax=max_val)
axis('off')
draw()
if save_plot:
savefig('../results/optimized_network.png')
show(block=blocking) # Open matplotlib window
def count_seq(x):
stop = np_nonzero(np_diff(np_hstack((x, 0))) == -1)
start = np_nonzero(np_diff(np_hstack((0, x))) == 1)
ld = {}
if len(start[0]):
d = stop[0] - start[0] + 1
for ll in xrange(np_max((np_min(d), 2)), np_max(d) + 1):
ld[str(ll)] = np_sum((d == ll).astype(int))
return ld
def guessGran(massAcid, emf, tempK, massSample, concAcid):
"""Simple Gran plot first guesses for alkalinity and EMF0."""
f1Guess = f1(massAcid, emf, tempK, massSample)
LGuess = logical_and(
f1Guess > 0.1*np_max(f1Guess),
f1Guess < 0.9*np_max(f1Guess),
)
alkGuess = granAlkGuess(massAcid[LGuess], f1Guess[LGuess], massSample,
concAcid)
emf0Guess = mean(granEmf0Guess(massAcid[LGuess], emf[LGuess],
tempK[LGuess], massSample, concAcid, alkGuess))
hGuess = emf2h(emf, emf0Guess, tempK)
pHGuess = -log10(hGuess)
return alkGuess, emf0Guess, hGuess, pHGuess
def quality(data):
_z_ = logreg_predict(data)[0]
sample = X[ y==_z_ ]
sample_ = X[ y==(1-_z_) ]
d = np_linalg_norm( [data] - sample )
d_ = np_linalg_norm( [data] - sample_ )
r = np_max( d_ )/np_max( d )
# r = np_mean( d )/np_mean( d_ )
# r = np_min( d )/np_min( d_ )
# r = 0.5 * ( r + recall )
if r > 1:
r = abs( 1-r )
r = 0.5 * ( r + recall )
return str( r )
def get_arrows_plt(mesh_pv, field, meshsol, factor, is_point_arrow, phase=1):
"""Create a pyvista arrow plot
Parameters
----------
mesh_pv : UnstructuredGrid
a pyvista mesh object
field : ndarray
a vector field to plot as glyph
meshsol : MeshSolution
a MeshSolution object
factor : float
an amplitude factor for the glyph plot
is_point_arrow : bool
True to plot arrows on the nodes
phase : complex
a phase shift to apply on the plot
Returns
-------
arrows_plt : PolyData
a pyvista object to plot glyph
factor : float
an amplitude factor for the plot glyph
"""
vect_field = real(field * phase)
# Compute factor
if factor is None:
factor = 0.2 * np_max(np_abs(mesh_pv.bounds)) / np_max(np_abs(vect_field))
# Add third dimension if needed
solution = meshsol.get_solution()
if solution.dimension == 2:
vect_field = hstack((vect_field, zeros((vect_field.shape[0], 1))))
# Add field to mesh
if is_point_arrow:
mesh_pv.vectors = vect_field * factor
arrows_plt = mesh_pv.arrows
else:
mesh_pv["field"] = vect_field
mesh_cell = mesh_pv.point_data_to_cell_data()
surf = mesh_cell.extract_geometry()
centers2 = surf.cell_centers()
centers2.vectors = surf["field"] * factor
arrows_plt = centers2.arrows
return arrows_plt, factor
def prep(datFile,
volSample,
pSal,
totalCarbonate,
totalPhosphate,
totalSilicate,
concAcid,
WhichKs=10,
WhoseKSO4=1,
WhoseKF=1,
WhoseTB=2,
totalAmmonia=0,
totalH2Sulfide=0):
"""Preparatory calculations for plotting."""
massAcid, emf, tempK, massSample, concTotals, eqConstants = \
datfile.prep(datFile, volSample, pSal, totalCarbonate, totalPhosphate,
totalSilicate, WhichKs=WhichKs, WhoseKSO4=WhoseKSO4, WhoseKF=WhoseKF,
WhoseTB=WhoseTB, totalAmmonia=totalAmmonia,
totalH2Sulfide=totalH2Sulfide)
f1Guess = solve.f1(massAcid, emf, tempK, massSample)
LGuess = logical_and(
f1Guess > 0.1 * np_max(f1Guess),
f1Guess < 0.9 * np_max(f1Guess),
)
alkGuess, emf0Guess, _, pHGuess = solve.guessGran(massAcid, emf, tempK,
massSample, concAcid)
granEmf0 = solve.granEmf0Guess(massAcid[LGuess], emf[LGuess],
tempK[LGuess], massSample, concAcid,
alkGuess)
L = logical_and(pHGuess > 3, pHGuess < 4)
alk_emf0 = solve.complete(massAcid, emf, tempK, massSample, concAcid,
concTotals, eqConstants)
h = solve.emf2h(emf, alk_emf0['x'][1], tempK)
mu = solve.mu(massAcid, massSample)
alkSim = simulate.alk(h, mu, concTotals, eqConstants)
alk0Sim = (alkSim[0] + massAcid * concAcid / (massAcid + massSample)) / mu
RMS = sqrt(mean(alk_emf0['fun']**2))
Npts = len(alk_emf0['fun'])
rgb = zeros((len(emf), 3))
for i in range(len(rgb)):
if LGuess[i] and L[i]:
rgb[i] = _rgb_both
elif LGuess[i] and not L[i]:
rgb[i] = _rgb_guess
elif L[i] and not LGuess[i]:
rgb[i] = _rgb_final
else:
rgb[i] = array([1, 1, 1])
return (massAcid, emf, massSample, f1Guess, LGuess, alkGuess, emf0Guess,
granEmf0, alk_emf0, alkSim, alk0Sim, RMS, Npts, rgb)
def mmr_geometricmedian(X):
(m,n)=X.shape
u=mean(X,axis=0)
niter=1000
xeps=sqrt(np_sum(u**2))/1000
xerr=2*xeps
for i in range(niter):
d2u=sqrt(np_sum((X-tile(u,(m,1)))**2,axis=1))
inul=where(d2u<xeps)[0]
d2u[inul]=xeps
unext=np_sum(X/tile(d2u.reshape((m,1)),(1,n)),axis=0)/np_sum(ones(m)/d2u)
if np_max(unext-u)<xerr:
break
u=copy(unext)
return(unext,i,np_max(unext-u))
def comp_point_ref(self, is_set=False):
"""Compute the point ref of the Surface
Parameters
----------
self : SurfLine
A SurfLine object
is_set: bool
True to update the point_ref property
Returns
-------
point_ref : complex
the reference point of the surface
"""
middle_array = array([line.get_middle() for line in self.get_lines()])
point_ref = sum(middle_array) / middle_array.size
# Use another method if the point is not is the surface
if not self.is_inside(Z=point_ref, if_online=False):
middle_array_abs = abs(middle_array)
# Find "min abs" middle
mid_id = argmin(middle_array_abs)
Zmid = middle_array[mid_id]
H = np_max(middle_array_abs) - np_min(middle_array_abs)
point_ref = (abs(Zmid) + H / 100) * exp(1j * angle(Zmid))
if is_set:
self.point_ref = point_ref
return point_ref
def update(self, abs_tol=1e-5, rel_tol=1e-3):
'''
update should return a number that when it is smaller than 1
the main loop stops. Here I choose this number to be:
sqrt(1/dim*sum_{i=0}^{dim}(grad/(abs_tol+x*rel_tol))_i^2)
'''
# accumulate the decay rates, in order to correct the averages
self.beta_m_ac *= self.beta_m_ac
_w2 = 0
_check = 0
self.Q.averageGrad()
for i in range(self.dim):
self.mE[i] = self.beta_m * self.mE[i] + (
1 - self.beta_m) * self.Q.grad[i]
self.v_max[i] = np_max(
[self.beta_v * self.v_max[i],
np_abs(self.Q.grad[i])])
dw = self.alpha / (self.v_max[i] + self.epsilon) * self.mE[i] / (
1 - self.beta_m_ac)
self.Q.model.w[i] = self.Q.model.w[i] - dw
_w2 = abs_tol + np_abs(self.Q.model.w[i]) * rel_tol
_check += (dw / _w2) * (dw / _w2)
self.Q.grad[i] = 0
_check = np_sqrt(1. / self.dim * _check)
return _check
def get_field(self, axes_list):
"""Returns the values of the field (with symmetries and sums).
Parameters
----------
self: Data
a Data object
axes_list: list
a list of RequestedAxis objects
Returns
-------
values: ndarray
values of the field
"""
values = self.values
for axis_requested in axes_list:
# Rebuild symmetries when needed
axis_symmetries = self.axes[axis_requested.index].symmetries
if (
axis_requested.transform == "fft"
and axis_requested.is_pattern
or axis_requested.extension in ["sum", "rss", "mean", "rms", "integrate"]
and axis_requested.is_pattern
):
values = take(values, axis_requested.rebuild_indices, axis_requested.index)
elif axis_requested.transform == "fft" and "antiperiod" in axis_symmetries:
nper = axis_symmetries["antiperiod"]
axis_symmetries["antiperiod"] = 2
values = rebuild_symmetries(values, axis_requested.index, axis_symmetries)
axis_symmetries["antiperiod"] = nper
elif axis_requested.indices is not None:
if (
axis_requested.extension in ["sum", "rss", "mean", "rms", "integrate"]
or max(axis_requested.indices) > values.shape[axis_requested.index]
):
values = rebuild_symmetries(
values, axis_requested.index, axis_symmetries
)
self.axes[axis_requested.index].symmetries = dict()
# sum over sum axes
if axis_requested.extension == "sum":
values = np_sum(values, axis=axis_requested.index, keepdims=True)
# root sum square over rss axes
elif axis_requested.extension == "rss":
values = sqrt(np_sum(values ** 2, axis=axis_requested.index, keepdims=True))
# mean value over mean axes
elif axis_requested.extension == "mean":
values = np_mean(values, axis=axis_requested.index, keepdims=True)
# RMS over rms axes
elif axis_requested.extension == "rms":
values = sqrt(
np_mean(values ** 2, axis=axis_requested.index, keepdims=True)
)
# integration over integration axes
elif axis_requested.extension == "integrate":
values = trapz(
values, x=axis_requested.values, axis=axis_requested.index
) / (np_max(axis_requested.values) - np_min(axis_requested.values))
return values
def mmr_geometricmedian_ker(K):
m = K.shape[0]
Ka = mean(K, axis=1)
niter = 1000
xeps = sqrt(np_sum(Ka**2)) / 1000
xerr = 2 * xeps
Q = linalg.qr(K)[0]
Ra = dot(Q.T, Ka)
aKa = dot(Ra, Ra)
for i in range(niter):
d2u = sqrt((zeros(m) + aKa) + diag(K) - 2 * Ka)
inul = where(d2u < xeps)[0]
d2u[inul] = xeps
Kanext=np_sum(K/tile(d2u.reshape((m,1)),(1,m)),axis=0) \
/np_sum(ones(m)/d2u)
du = np_sum((K - tile(Ka, (m, 1))).T / tile(d2u, (m, 1)), axis=1)
print(sqrt(np_sum(du**2)))
if np_max(Kanext - Ka) < xerr:
Ka = copy(Kanext)
Ra = dot(Q.T, Ka)
aKa = dot(Ra, Ra)
break
Ka = copy(Kanext)
Ra = dot(Q.T, Ka)
aKa = dot(Ra, Ra)
return (Ka, aKa)
def get_fit(self,
fit_type,
data,
order,
smooth,
degree,
begin,
end,
weight=None):
z1 = np_zeros(begin)
z2 = np_zeros(len(data[0]) - end)
if end == 0:
end = None
x = data[0][begin:end]
y = data[1][begin:end]
if weight is not None:
weight = weight[begin:end]
if fit_type == "spline":
f = inter.UnivariateSpline(x, y, w=weight, k=order, s=smooth)
else:
f = poly.Chebyshev.fit(x, y, degree, w=weight)
res = y - f(x)
nfit = f(x) / np_max(f(x))
corr = concatenate((z1, nfit, z2))
fitc = [x, f(x), corr, res]
return fitc
def maximum_spread(self,
pareto_front=None,
reference_front=None): ## MS function
""" It addresses the range of objective function values and takes into account the proximity to the true Pareto front"""
pareto_front, reference_front = self.get_pareto_front_reference_front(
pareto_front, reference_front)
n_objs = reference_front.shape[1]
pf_max = np_max(pareto_front, axis=0)
pf_min = np_min(pareto_front, axis=0)
rf_max = np_max(reference_front, axis=0)
rf_min = np_min(reference_front, axis=0)
ms = 0
for i in range(0, n_objs):
ms += ((min(pf_max[i], rf_max[i]) - max(pf_min[i], rf_min[i])) /
(rf_max[i] - rf_min[i]))**2
return sqrt(ms / n_objs)
def plot_animated(self, i):
"""Return the pyvista mesh object.
Parameters
----------
self : Mode
a Mode object
i : int
index of the mode to plot
Returns
-------
mesh : pyvista.core.pointset.StructuredGrid
a pyvista StructuredGrid object
"""
radial_shape = self.get_shape_pol()[:, 0]
shape_xyz = self.get_shape_xyz()
clim = [np_min(radial_shape), np_max(radial_shape)]
self.parent.mesh.plot_deformation_animated(
shape_xyz,
radial_shape,
factor=0.05,
field_name="Radial displacement",
clim=clim,
)
def mmr_geometricmedian_ker(K):
m=K.shape[0]
Ka=mean(K,axis=1)
aKa=np_sum(Ka)/m
niter=1000
xeps=sqrt(np_sum(Ka**2))/100
xerr=2*xeps
e1=ones(m)
for iiter in range(niter):
## d2u=sqrt((zeros(m)+aKa)+diag(K)-2*Ka)
d2u_2=aKa+diag(K)-2*Ka
ineg=where(d2u_2<0)[0]
d2u_2[ineg]=0.0
d2u=sqrt(d2u_2)
inul=where(d2u<xeps)[0]
d2u[inul]=xeps
xdenom=np_sum(e1/d2u)
Kanext=np_sum(K/outer(d2u,e1),axis=0)/xdenom
aKanext=np_sum(Ka/d2u)/xdenom
if np_max(Kanext-Ka)<xerr:
Ka=copy(Kanext)
aKa=aKanext
break
Ka=copy(Kanext)
aKa=aKanext
return(Ka,aKa)
def components(ax, massAcid, alk0Sim, alkSim, sublabel):
"""Every component of alkalinity throughout a titration."""
ax.plot(massAcid * 1e3,
-log10(alk0Sim),
label='Total alk.',
marker='o',
markersize=3,
c='k',
clip_on=False)
for component in alkSim[1].keys():
if component.startswith('-'): # this is a bit sketchy
yVar = -alkSim[1][component]
else:
yVar = alkSim[1][component]
if np_any(yVar > 0):
ax.plot(massAcid * 1e3,
-log10(yVar),
label=rgbs_names[component][1],
marker='x',
markersize=3,
c=rgbs_names[component][0],
clip_on=False)
ax.set_xlim([0, np_max(massAcid) * 1e3])
ax.set_ylim(ax.get_ylim()[::-1])
ax.legend(bbox_to_anchor=(1.05, 1))
ax.set_xlabel('Acid mass / g')
ax.set_ylabel('$-$log$_{10}$(concentration from pH / mol$\cdot$kg$^{-1}$)')
ax.set_title(sublabel, fontsize=10)
return ax
def mmr_geometricmedian_ker(K):
m=K.shape[0]
Ka=mean(K,axis=1)
aKa=np_sum(Ka)/m
niter=1000
xeps=sqrt(np_sum(Ka**2))/100
xerr=2*xeps
e1=ones(m)
for iiter in range(niter):
## d2u=sqrt((zeros(m)+aKa)+diag(K)-2*Ka)
d2u_2=aKa+diag(K)-2*Ka
ineg=where(d2u_2<0)[0]
d2u_2[ineg]=0.0
d2u=sqrt(d2u_2)
inul=where(d2u<xeps)[0]
d2u[inul]=xeps
xdenom=np_sum(e1/d2u)
Kanext=np_sum(K/outer(d2u,e1),axis=0)/xdenom
aKanext=np_sum(Ka/d2u)/xdenom
if np_max(Kanext-Ka)<xerr:
Ka=copy(Kanext)
aKa=aKanext
break
Ka=copy(Kanext)
aKa=aKanext
return(Ka,aKa)
def train(self):
pop = [self.create_solution() for _ in range(self.pop_size)]
v_max = 0.5 * (self.ub - self.lb)
v_min = zeros(self.problem_size)
v_list = uniform(v_min, v_max, (self.pop_size, self.problem_size))
pop_local = deepcopy(pop)
g_best = self.get_global_best_solution(pop=pop,
id_fit=self.ID_FIT,
id_best=self.ID_MIN_PROB)
N_CLS = int(self.pop_size / 5) # Number of chaotic local searches
for epoch in range(self.epoch):
r = rand()
list_fits = [item[self.ID_FIT] for item in pop]
fit_avg = mean(list_fits)
fit_min = np_min(list_fits)
for i in range(self.pop_size):
w = self.__get_weights__(pop[i][self.ID_FIT], fit_avg, fit_min)
v_new = w * v_list[i] + self.c1 * rand() * (pop_local[i][self.ID_POS] - pop[i][self.ID_POS]) + \
self.c2 * rand() * (g_best[self.ID_POS] - pop[i][self.ID_POS])
x_new = pop[i][self.ID_POS] + v_new
x_new = self.amend_position_random_faster(x_new)
fit_new = self.get_fitness_position(x_new)
pop[i] = [x_new, fit_new]
# Update current position, current velocity and compare with past position, past fitness (local best)
if fit_new < pop_local[i][self.ID_FIT]:
pop_local[i] = [x_new, fit_new]
g_best = self.update_global_best_solution(pop, self.ID_MIN_PROB,
g_best)
## Implement chaostic local search for the best solution
cx_best_0 = (g_best[self.ID_POS] - self.lb) / (self.ub - self.lb
) # Eq. 7
cx_best_1 = 4 * cx_best_0 * (1 - cx_best_0) # Eq. 6
x_best = self.lb + cx_best_1 * (self.ub - self.lb) # Eq. 8
fit_best = self.get_fitness_position(x_best)
if fit_best < g_best[self.ID_FIT]:
g_best = [x_best, fit_best]
bound_min = stack(
[self.lb, g_best[self.ID_POS] - r * (self.ub - self.lb)])
self.lb = np_max(bound_min, axis=0)
bound_max = stack(
[self.ub, g_best[self.ID_POS] + r * (self.ub - self.lb)])
self.ub = np_min(bound_max, axis=0)
pop_new_child = [
self.create_solution() for _ in range(self.pop_size - N_CLS)
]
pop_new = sorted(pop, key=lambda item: item[self.ID_FIT])
pop = pop_new[:N_CLS] + pop_new_child
self.loss_train.append(g_best[self.ID_FIT])
if self.verbose:
print(">Epoch: {}, Best fit: {}".format(
epoch + 1, g_best[self.ID_FIT]))
self.solution = g_best
return g_best[self.ID_POS], g_best[self.ID_FIT], self.loss_train
def alkComponents(titrationPotentiometric, ax=None):
t = titrationPotentiometric
assert 'alkSteps' in vars(t), \
'You must first run `titrationPotentiometric.get_alkSteps().`'
# Get the 'used' values
solver = 'complete' # only valid option for now
usedMin = np_min(t.volAcid[t.solvedWith[solver]])
usedMax = np_max(t.volAcid[t.solvedWith[solver]])
# Draw the plot
ax = _checksetax(ax)
ax.plot(t.volAcid,
-log10(t.alkSteps),
label='Total alk.',
marker='o',
markersize=_markersize,
c='k',
alpha=_alpha)
for component, conc in t.alkComponents.items():
if np_any(conc != 0):
ax.plot(t.volAcid, -log10(np_abs(conc)), **rgbs[component])
ax.add_patch(
patches.Rectangle((usedMin, ax.get_ylim()[1]),
usedMax - usedMin,
ax.get_ylim()[0] - ax.get_ylim()[1],
facecolor=0.9 * ones(3)))
ax.invert_yaxis()
ax.legend(bbox_to_anchor=(1.05, 1), edgecolor='k')
ax.set_xlabel('Acid volume / ml')
ax.set_ylabel('$-$log$_{10}$(concentration from pH / mol$\cdot$kg$^{-1}$)')
return ax
def statistic(float_set):
"""
This function calculates standard statistical metrics over a list of floats.
:param float_set: list of float
:return: Dictionary of float and list of float
Statistic of the set of instances {mean:, median:, min:, max:, sample_variance:, samples:}
"""
float_set_size = len(float_set)
# Calculate the sample mean
factor = 1 / float_set_size
sample_mean = factor * np_sum(float_set)
# Calculate the standard deviation
factor = 1 / (float_set_size - 1)
sample_variance = sqrt(factor * np_sum((float_set - sample_mean)**2))
minimum = np_min(float_set)
maximum = np_max(float_set)
median_dist = median(float_set)
return {
'mean': sample_mean,
'median': median_dist,
'min': minimum,
'max': maximum,
'sv': sample_variance,
'samples': float_set
}
def solve_FEMM(self, output, sym, FEMM_dict):
# Loading parameters for readibility
angle = output.mag.angle
qs = output.simu.machine.stator.winding.qs # Winding phase number
Npcpp = output.simu.machine.stator.winding.Npcpp
L1 = output.simu.machine.stator.comp_length()
Nt_tot = output.mag.Nt_tot # Number of time step
Na_tot = output.mag.Na_tot # Number of angular step
# Create the mesh
femm.mi_createmesh()
# Initialize results matrix
Br = zeros((Nt_tot, Na_tot))
Bt = zeros((Nt_tot, Na_tot))
Tem = zeros((Nt_tot, 1))
Phi_wind_stator = zeros((Nt_tot, qs))
# Compute the data for each time step
for ii in range(Nt_tot):
# Update rotor position and currents
update_FEMM_simulation(
output,
FEMM_dict["materials"],
FEMM_dict["circuits"],
self.is_mmfs,
self.is_mmfr,
j_t0=ii,
)
# Run the computation
femm.mi_analyze()
femm.mi_loadsolution()
# Get the flux result
for jj in range(Na_tot):
Br[ii, jj], Bt[ii, jj] = femm.mo_getgapb("bc_ag2",
angle[jj] * 180 / pi)
# Compute the torque
Tem[ii] = comp_FEMM_torque(FEMM_dict, sym=sym)
# Phi_wind computation
Phi_wind_stator[ii, :] = comp_FEMM_Phi_wind(qs,
Npcpp,
is_stator=True,
Lfemm=FEMM_dict["Lfemm"],
L1=L1,
sym=sym)
# Store the results
output.mag.Br = Br
output.mag.Bt = Bt
output.mag.Tem = Tem
output.mag.Tem_av = mean(Tem)
if output.mag.Tem_av != 0:
output.mag.Tem_rip = abs(
(np_max(Tem) - np_min(Tem)) / output.mag.Tem_av)
output.mag.Phi_wind_stator = Phi_wind_stator
# Electromotive forces computation (update output)
self.comp_emf()
def halfGran(massAcid, emf, tempK, massSample, concAcid, concTotals,
eqConstants, pHRange=[3., 4.], suppressWarnings=False):
"""Solve for alkalinity and EMF0 using the half-Gran method [H15]."""
xmu = mu(massAcid, massSample)
granReps = int(20)
stepAlk = full(granReps, nan)
stepEmf0 = full(granReps, nan)
granHSO4 = zeros(size(emf))
granHF = zeros(size(emf))
granG = full((granReps, size(emf)), nan)
granH = full((granReps, size(emf)), nan)
granEmf0 = full((granReps, size(emf)), nan)
granPH = full((granReps, size(emf)), nan)
converged = False
for i in range(granReps):
if i == 0:
granG[i] = f1(massAcid, emf, tempK, massSample)
LG = granG[i] > 0.1*np_max(granG[i])
else:
LG = logical_and(granPH[i-1] > pHRange[0],
granPH[i-1] < pHRange[1])
stepAlk[i] = granAlkGuess(massAcid[LG], granG[i, LG], massSample,
concAcid)
PPC = 5e-3 # permitted % change in AT
if i > 2:
if np_abs(stepAlk[i] - stepAlk[i-1])/stepAlk[i] < PPC/100:
converged = True
break
granEmf0[i, LG] = granEmf0Guess(massAcid[LG], emf[LG], tempK[LG],
massSample, concAcid, stepAlk[i], granHSO4[LG], granHF[LG])
stepEmf0[i] = nanmean(granEmf0[i])
granH[i] = emf2h(emf, stepEmf0[i], tempK)
granPH[i] = -log10(granH[i])
granBicarb = xmu*concTotals['C']/(granH[i]/eqConstants['C1'] + 1)
granHSO4 = xmu*concTotals['S']/(1 + eqConstants['S']/granH[i])
granHF = xmu*concTotals['F']/(1 + eqConstants['F']/granH[i])
granBorate = xmu*concTotals['B']/(1 + granH[i]/eqConstants['B'])
granOH = eqConstants['w']/granH[i]
granPP2 = (xmu*concTotals['P']*(1 -
eqConstants['P1']*eqConstants['P2']/(granH[i]**2))
/ (1 + eqConstants['P1']/granH[i] +
eqConstants['P2']*eqConstants['P3']/granH[i]**2 +
eqConstants['P1']*eqConstants['P2']*eqConstants['P3']/granH[i]**3))
if i < granReps-1:
granG[i+1] = (granH[i] + granHSO4 + granHF - granBicarb
- granOH - granBorate + granPP2)*(massSample + massAcid)
if converged:
finalAlk = stepAlk[~isnan(stepAlk)][-1]
finalEmf0 = stepEmf0[~isnan(stepEmf0)][-1]
else:
if not suppressWarnings:
print('Calkulate: half-Gran plot iterations did not converge!')
finalAlk = nan
finalEmf0 = nan
optResult = {
'x': [finalAlk, finalEmf0],
'L': LG,
}
return optResult
def plot_hsm(ax, result, wavelengths):
"""
Plotter of a HSM result
----------------------
:param ax: ax object to edit
:param result: result to plot
:param wavelengths: wavelengths used to get result. Will become x-axis
:return: None. Edits ax object
"""
def lorentzian(params, x):
"""
Lorentzian formula. Taken from SPectrA
----------------
:param params: Parameters of lorentzian. Need to be four.
:param x: x-axis. Wavelengths
:return: array of values for current parameters and wavelengths
"""
return params[0] + params[1] / ((x - params[2])**2 +
(0.5 * params[3])**2)
# scatter plot of actual results
ax.scatter(wavelengths, result['intensity'])
try:
# try to fit and plot fit over
wavelengths_ev = 1240 / linspace(
np_min(wavelengths), np_max(wavelengths), num=50)
hsm_fit = lorentzian(result['fit_parameters'], wavelengths_ev)
ax.plot(1240 / wavelengths_ev, hsm_fit, 'r--')
except:
pass
# set axis to same range every time
ax.set_xlim(np_min(wavelengths) - 30, np_max(wavelengths) + 30)
try:
# add fit info if possible
text = '\n'.join((r'SPR (nm)=%.1f' % (result['result'][0], ),
r'linewidth (meV)=%.1f' % (result['result'][1], ),
r'r^2=%.1f' % (result['result'][2], )))
props = dict(boxstyle='round', facecolor='white', alpha=0.5)
ax.text(0.05,
0.95,
text,
transform=ax.transAxes,
verticalalignment='top',
bbox=props)
except:
pass
def norm(array):
from numpy import nanmin as np_min
from numpy import nanmax as np_max
norm_array = (array - np_min(array)) / (np_max(array) - np_min(array))
return norm_array
def log_datakeeper_step_result(simulation, datakeeper_list, index, simu_type):
"""Log the content of the datakeeper for the step index (if index=None, use reference)"""
if simulation.layer == 2:
msg = " "
else:
msg = ""
if simu_type is not None:
msg += simu_type + " "
if simulation.index is None:
msg += "Reference "
msg += "Results: "
for datakeeper in datakeeper_list:
if index is None:
value = datakeeper.result_ref
else:
value = datakeeper.result[index]
# Format log
if isinstance(value, ndarray):
msg += (datakeeper.symbol + "=array(min=" +
format(np_min(value), ".4g") + ",max=" +
format(np_max(value), ".4g") + ")")
elif isinstance(value, list):
msg += (datakeeper.symbol + "=list(min=" +
format(np_min(value), ".4g") + ",max=" +
format(np_max(value), ".4g") + ")")
elif isinstance(value, Data) or isinstance(value, VectorField):
msg += datakeeper.symbol + "=" + type(value).__name__
elif value is None:
pass
# msg += datakeeper.symbol + "=None"
else:
msg += datakeeper.symbol + "=" + format(value, ".4g")
if value is not None:
if datakeeper.unit is not None:
msg += " [" + datakeeper.unit + "], "
else:
msg += ", "
msg = msg[:-2]
# Get logger of the main simulation in parallel mode
if simulation.logger_name[0:8] == "Parallel":
log = getLogger(simulation.logger_name[9:])
else:
log = simulation.get_logger()
log.info(msg)
def deduplicate_and_interpolate(x, y, **kwargs):
"""
As interp1d has issues with duplicates, fix x and y so that it works
"""
dedup_dict = defaultdict(list)
for x_i, y_i in zip(x, y):
dedup_dict[x_i].append(y_i)
fixed_x, fixed_y = zip(*[(x_i, np_max(y_i))
for x_i, y_i in dedup_dict.items()])
return interp1d(fixed_x, fixed_y, **kwargs)
def add_curve(self, curve):
"""
Add a curve to the bounding box.
:param curve: Curve to add to bounding box.
:type curve: :class:`.BezierCurve` or :class:`.NurbsCurve`
"""
cp = curve.cp
bmin = np_min(cp, axis=0)
bmax = np_max(cp, axis=0)
self.set_bounds(bmin, bmax)
def show_entropy(xyz, H):
H_color = (H - np_min(H)) / (np_max(H) - np_min(H))
fig = figure()
ax = Axes3D(fig)
xyz_b = xyz[where(H_color <= 0.25)[0]]
xyz_r = xyz[where(H_color > 0.75)[0]]
print(np_min(H), np_max(H), H[1], H_color[1], H_color, xyz_b.shape,
xyz_r.shape)
# xyz_g = xyz[where((H_color>0.25) and (H_color<=0.75))[0]]
# ax.scatter(xyz[:,0],xyz[:,1],xyz[:,2], c=H_color)
ax.scatter(xyz_b[:, 0], xyz_b[:, 1], xyz_b[:, 2], color='b')
# ax.scatter(xyz_g[:,0],xyz_g[:,1],xyz_g[:,2], color='g')
ax.scatter(xyz_r[:, 0], xyz_r[:, 1], xyz_r[:, 2], color='r')
ax.set_xticks(arange(-1, 1, 0.2))
ax.set_yticks(arange(-1, 1, 0.2))
ax.set_zticks(arange(-1, 1, 0.2))
grid()
fig.show()
def f1Gran(ax, massAcid, massSample, concAcid, f1Guess, alkGuess, rgb,
sublabel):
"""F1 Gran plot function for the first alkalinity estimate."""
ax.axvline(1e3 * alkGuess * massSample / concAcid,
color=_rgb_guess,
linestyle='--',
zorder=1)
ax.scatter(massAcid * 1e3,
f1Guess * 1e-7,
c=rgb,
edgecolors='k',
clip_on=False,
zorder=2)
ax.set_xlim([0, np_max(massAcid) * 1e3])
ax.set_ylim([0, np_max(f1Guess * 1.05e-7)])
ax.set_xlabel('Acid mass / g')
ax.set_ylabel('$F_1 \cdot 10^{-7}$')
ax.set_title('{} First-guess alkalinity = {:.1f} μmol/kg'.format(
sublabel, alkGuess * 1e6),
fontsize=10)
return ax
def draw_concentric_contours(steps,
text,
textposition,
title,
filename,
plotly_data=[],
steps_mode='lines, text, markers',
debug_print=False):
ply_data = []
if plotly_data is not None: ply_data.extend(plotly_data)
ams = []
if steps is None and debug_print:
print("No steps")
elif debug_print:
print("{}: steepest-descent: start at x0={}".format(title, x0))
print("number of steps for Steepest Descent={}".format(steps.shape[1]))
sef = True
for k, ec in zip(ks, ecs):
d, am, _ = qfc.level_k_ellipsoid(A,
b,
c,
alpha,
beta,
k=k,
ellipsoid_color=ec,
show_eigenvectors=sef,
debug_print=debug_print)
sef = False
ply_data.extend(d)
ams.append(am)
if steps is not None:
steps_ply_data = Scatter_R2(steps[0],
steps[1],
color='rgb(0,0,140)',
width=1.,
mode=steps_mode,
text=text,
textposition=textposition,
textfontsize=18,
hoverinfo='x+y')
ply_data.append(steps_ply_data)
axes_max = np_max(np_abs(ams))
plot_it_R2(ply_data,
axes_max,
title=title,
filename=filename,
buffer_scale=1.,
buffer_fixed=.1)
return
def plot_density(data, smooth = False):
min = 0
max = np_max(data)
data = np.ravel(data)
count = np.bincount(data, minlength=max + 1)
nums = np.arange(min, max+1, 1)
if smooth:
nums_new = np.linspace(min, max, 1000)
count_new = spline(nums, count, nums_new)
fig = plt.figure()
if smooth:
plt.plot(nums_new, count_new)
else:
plt.plot(nums, count)
plt.show()
def makeColourProfile(self):
"""Make a colour profile based on ksig information"""
working_data = np_array(self.kmerSigs, copy=True)
Center(working_data,verbose=0)
p = PCA(working_data)
components = p.pc()
# now make the colour profile based on PC1
self.kmerVals = np_array([float(i) for i in components[:,0]])
# normalise to fit between 0 and 1
self.kmerVals -= np_min(self.kmerVals)
self.kmerVals /= np_max(self.kmerVals)
if(False):
plt.figure(1)
plt.subplot(111)
plt.plot(components[:,0], components[:,1], 'r.')
plt.show()
label = Base[0 : Base.find(".txt")].replace("_", " ")
n_Components = len(Components)
Wmin_vector = zeros(n_Components)
Wmax_vector = zeros(n_Components)
for j in range(len(Components)):
Wavelength = pv.get_ColumnData(
[0],
Components_Folder + Components[j],
HeaderSize=0,
StringIndexes=False,
comments_icon="#",
unpack_check=True,
)
Wmin_vector[i], Wmax_vector[i] = np_min(Wavelength), np_max(Wavelength)
Axis3D.bar3d(log10(Age), metallicity, Wmin_vector, ones(len(Age)), ones(len(metallicity)), Wmax_vector)
# Axis3D.xaxis.set_scale('log')
# Axis3D.xaxis.set_xlim(100000, 5e10)
plt.show()
# pv.DataPloter_One(Age, metallicity, label, pv.Color_Vector[2][i+2], LineStyle=None)
# Plot_Title = 'Stellar bases used in SSP synthesis'
# Plot_ylabel = 'Metallicity'
# Plot_xlabel = 'Age ' + r'$(yr)$'
#
# pv.Labels_Legends_One(Plot_Title, Plot_xlabel, Plot_ylabel, LegendLocation = 'best')
def mmr_normalization(ilocal,iscale,XTrain,XTest,ipar):
## function to normalize the input and the output data
## !!!! the localization happens before normalization if both given !!!
## input
## ilocal centralization
## =-1 no localization
## =0 mean
## =1 median
## =2 geometric median
## =3 shift by ipar
## =4 smallest enclosing ball
## =5 row mean row wise
## icenter
## =-1 no scaling
## =0 scale item wise by L2 norm
## =1 scale item wise by L1 norm
## =2 scale item wise by L_infty norm
## =3 scale items by stereographic projection relative to zero
## =4 scale variables by STD(standard deviation)
## =5 scale variables by MAD(median absolute deviation)
## =6 scale variables by absolute deviation
## =7 scale all variables by average STD
## =8 scale all variables by maximum STD
## =9 scale all variables by median MAD
## =10 scale item wise by Minkowski norm, power given by ipar
## =11 \sum_i||u-x_i||/m where u=0
## =12 scale all variables by overall max
## =13 Mahalonobis scaling
## XTrain Data matrix which will be normalized. It assumed the
## rows are the sample vetors and the columns are variables
## XTest Data matrix which will be normalized. It assumed the
## rows are the sample vetors and the columns are
## variables.
## It herites the center and the scale in the variable wise ca## se from the XTrain,
## otherwise it is normalized independently
## ipar additional parameter
## output
## XTrain Data matrix which is the result of the normalization
## of input XTrain. It assumed the rows are the sample
## vetors and the columns are variables
## XTest Data matrix which is the result of the normalization
## of input XTest. It assumed the rows are the sample
## vetors and the columns are variables.
## opar the radius in case of ixnorm=2.
##
if XTest is None:
XTest=array([])
opar=0;
(mtrain,n)=XTrain.shape
if len(XTest.shape)>=2:
mtest=XTest.shape[0]
elif len(XTest.shape)==1:
mtest=XTest.shape[0]
XTest=XTest.reshape((mtest,1))
else:
mtest=0
XTest=array([])
if ilocal==-1:
pass
elif ilocal==0: ## mean
xcenter=mean(XTrain,axis=0)
elif ilocal==1: ## median
xcenter=median(XTrain,axis=0)
elif ilocal==2: ## geometric median
xcenter=mmr_geometricmedian(XTrain)[0]
elif ilocal==3: ## shift by ipar
xcenter=ipar
elif ilocal==4: ## smallest comprising ball
xalpha=mmr_outerball(0,XTrain)
xcenter=dot(XTrain.T,xalpha)
elif ilocal==5: ## row mean row wise
xcenter=mean(XTrain,axis=1)
if ilocal in (0,1,2,3,4):
XTrain=XTrain-tile(xcenter,(mtrain,1))
if mtest>0:
XTest=XTest-tile(xcenter,(mtest,1))
elif ilocal==5:
XTrain=XTrain-outer(xcenter,ones(n))
if mtest>0:
xcenter=mean(XTest,axis=1)
XTest=XTest-outer(xcenter,ones(n))
## itemwise normalizations
if iscale==-1:
pass
elif iscale==0: ## scale items by L2 norm
xscale_tra=sqrt(np_sum(XTrain**2,axis=1))
if mtest>0:
xscale_tes=sqrt(np_sum(XTest**2,axis=1))
elif iscale==1: ## scale items by L1 norm
xscale_tra=np_sum(abs(XTrain),axis=1)
if mtest>0:
xscale_tes=np_sum(abs(XTest),axis=1)
elif iscale==2: ## scale items by L_infty norm
xscale_tra=np_max(abs(XTrain),axis=1)
if mtest>0:
xscale_tes=np_max(abs(XTest),axis=1)
elif iscale==10: ## scale items by Minowski with ipar
xscale_tra=np_sum(abs(XTrain)**ipar,axis=1)**(1/ipar)
if mtest>0:
xscale_tes=np_sum(abs(XTest)**ipar,axis=1)**(1/ipar)
if iscale in (0,1,2,10):
xscale_tra=xscale_tra+(xscale_tra==0)
XTrain=XTrain/tile(xscale_tra.reshape(mtrain,1),(1,n))
if mtest>0:
xscale_tes=xscale_tes+(xscale_tes==0)
XTest=XTest/tile(xscale_tes.reshape(mtest,1),(1,n))
if iscale==3: ## scale items by stereographic projection relative to zero
xnorm2=np_sum(XTrain**2,axis=1)
R=ipar
xhom=ones(mtrain)/(xnorm2+R**2)
xhom2=xnorm2-R**2
XTrain=concatenate((2*R**2*XTrain*outer(xhom,ones(n)),R*xhom2*xhom), \
axis=1)
if mtest>0:
xnorm2=np_sum(XTest**2,axis=1)
xhom=ones(mtest)/(xnorm2+R**2)
xhom2=xnorm2-R**2
XTest=concatenate((2*R**2*XTest*outer(xhom,ones(n)),R*xhom2*xhom), \
axis=1)
## variable wise normalization relative to zero
## test has to use of the training scale
if iscale==-1:
pass
elif iscale==4: ## scale vars by std to zeros center
xscale=std(XTrain,axis=0)
## xscale=sqrt(mean(XTrain**2,axis=0))
elif iscale==5: ## scale vars by mad
xscale=median(abs(XTrain),axis=0)
elif iscale==6: ## scale vars by absolut deviation
xscale=mean(abs(XTrain),axis=0)
if iscale in (4,5,6):
xscale=xscale+(xscale==0)
XTrain=XTrain/tile(xscale,(mtrain,1))
if mtest>0:
XTest=XTest/tile(xscale,(mtest,1))
if iscale==-1:
pass
if iscale==7: ## scale vars by average std to zero center
## xscale=mean(std(XTrain,axis=0))
xscale=mean(sqrt(mean(XTrain**2,axis=0)))
elif iscale==8: ## scale vars by max std to zero center
## xscale=np_max(std(XTrain,axis=0))
xscale=np_max(sqrt(mean(XTrain**2,axis=0)))
elif iscale==9: ## scale vars by median mad
xscale=median(median(abs(XTrain),axis=0))
elif iscale==11: ## \sum_i||u-x_i||/m where u=0
xscale=mean(sqrt(np_sum(XTrain**2,axis=1)))
elif iscale==12: ## \sum_i||u-x_i||/m where u=0
xscale=XTrain.max()
## print(xscale)
if iscale in (7,8,9,11,12):
xscale=xscale+(xscale==0)
XTrain=XTrain/xscale
if mtest>0:
XTest=XTest/xscale
if iscale==13: ## scale by Mahalonobis
xsigma=dot(XTrain.T,XTrain) ## covariance
[w,v]=linalg.eigh(xsigma)
iw=where(w<=10**(-10))[0]
w[iw]=0.0
iw=where(w>0.0)[0]
w_sqinv=zeros(XTrain.shape[1])
w_sqinv[iw]=1/sqrt(w[iw])
XTrain=dot(XTrain,v)*outer(ones(mtrain),w_sqinv)
if mtest>0:
XTest=dot(XTest,v)*outer(ones(mtest),w_sqinv)
return(XTrain,XTest,opar)
def findNewClusterCenters(self, ss=0):
"""Find a putative cluster"""
inRange = lambda x, l, u: x >= l and x < u
# we work from the top view as this has the base clustering
max_index = np_argmax(self.blurredMaps[0])
max_value = self.blurredMaps[0].ravel()[max_index]
max_x = int(max_index / self.PM.scaleFactor)
max_y = max_index - self.PM.scaleFactor * max_x
max_z = -1
ret_values = [max_value, max_x, max_y]
start_span = int(1.5 * self.span)
span_len = 2 * start_span + 1
if self.debugPlots:
self.plotRegion(max_x, max_y, max_z, fileName="Image_" + str(self.imageCounter), tag="column", column=True)
self.imageCounter += 1
# make a 3d grid to hold the values
working_block = np_zeros((span_len, span_len, self.PM.scaleFactor))
# go through the entire column
(x_lower, x_upper) = self.makeCoordRanges(max_x, start_span)
(y_lower, y_upper) = self.makeCoordRanges(max_y, start_span)
super_putative_row_indices = []
for p in self.im2RowIndicies:
if inRange(p[0], x_lower, x_upper) and inRange(p[1], y_lower, y_upper):
for row_index in self.im2RowIndicies[p]:
# check that the point is real and that it has not yet been binned
if row_index not in self.PM.binnedRowIndicies and row_index not in self.PM.restrictedRowIndicies:
# this is an unassigned point.
multiplier = np_log10(self.PM.contigLengths[row_index])
self.incrementAboutPoint3D(
working_block, p[0] - x_lower, p[1] - y_lower, p[2], multiplier=multiplier
)
super_putative_row_indices.append(row_index)
# blur and find the highest value
bwb = ndi.gaussian_filter(working_block, 8) # self.blurRadius)
densest_index = np_unravel_index(np_argmax(bwb), (np_shape(bwb)))
max_x = densest_index[0] + x_lower
max_y = densest_index[1] + y_lower
max_z = densest_index[2]
# now get the basic color of this dense point
putative_center_row_indices = []
(x_lower, x_upper) = self.makeCoordRanges(max_x, self.span)
(y_lower, y_upper) = self.makeCoordRanges(max_y, self.span)
(z_lower, z_upper) = self.makeCoordRanges(max_z, 2 * self.span)
for row_index in super_putative_row_indices:
p = np_around(self.PM.transformedCP[row_index])
if inRange(p[0], x_lower, x_upper) and inRange(p[1], y_lower, y_upper) and inRange(p[2], z_lower, z_upper):
# we are within the range!
putative_center_row_indices.append(row_index)
# make sure we have something to go on here
if np_size(putative_center_row_indices) == 0:
# it's all over!
return None
if np_size(putative_center_row_indices) == 1:
# get out of here but keep trying
# the calling function may restrict these indices
return [[np_array(putative_center_row_indices)], ret_values]
else:
total_BP = sum([self.PM.contigLengths[i] for i in putative_center_row_indices])
if not self.isGoodBin(total_BP, len(putative_center_row_indices), ms=5): # Can we trust very small bins?.
# get out of here but keep trying
# the calling function should restrict these indices
return [[np_array(putative_center_row_indices)], ret_values]
else:
# we've got a few good guys here, partition them up!
# shift these guys around a bit
center_k_vals = np_array([self.PM.kmerVals[i] for i in putative_center_row_indices])
k_partitions = self.partitionVals(center_k_vals)
if len(k_partitions) == 0:
return None
else:
center_c_vals = np_array([self.PM.transformedCP[i][-1] for i in putative_center_row_indices])
# center_c_vals = np_array([self.PM.averageCoverages[i] for i in putative_center_row_indices])
center_c_vals -= np_min(center_c_vals)
c_max = np_max(center_c_vals)
if c_max != 0:
center_c_vals /= c_max
c_partitions = self.partitionVals(center_c_vals)
# take the intersection of the two partitions
tmp_partition_hash_1 = {}
id = 1
for p in k_partitions:
for i in p:
tmp_partition_hash_1[i] = id
id += 1
tmp_partition_hash_2 = {}
id = 1
for p in c_partitions:
for i in p:
try:
tmp_partition_hash_2[(tmp_partition_hash_1[i], id)].append(i)
except KeyError:
tmp_partition_hash_2[(tmp_partition_hash_1[i], id)] = [i]
id += 1
partitions = [
np_array([putative_center_row_indices[i] for i in tmp_partition_hash_2[key]])
for key in tmp_partition_hash_2.keys()
]
# pcs = [[self.PM.averageCoverages[i] for i in p] for p in partitions]
# print pcs
return [partitions, ret_values]
def _calculate_stats(self, xs, ys):
ag = self.analysis_group
ag.attribute = self.options.index_attr
ag.weighted_age_error_kind = self.options.error_calc_method
ag.include_j_error_in_mean = self.options.include_j_error_in_mean
ag.include_j_error_in_individual_analyses = self.options.include_j_error
mswd, valid_mswd, n = self.analysis_group.get_mswd_tuple()
if self.options.mean_calculation_kind == 'kernel':
wm, we = 0, 0
delta = 1
maxs, _mins = find_peaks(ys, xs, delta=delta, lookahead=1)
wm = np_max(maxs, axis=1)[0]
else:
wage = self.analysis_group.weighted_age
wm, we = wage.nominal_value, wage.std_dev
return wm, we, mswd, valid_mswd
# def _calc_error(self, we, mswd):
# ec = self.options.error_calc_method
# n = self.options.nsigma
# if ec == 'SEM':
# a = 1
# elif ec == 'SEM, but if MSWD>1 use SEM * sqrt(MSWD)':
# a = 1
# if mswd > 1:
# a = mswd ** 0.5
# return we * a * n
# ============= EOF =============================================
# def _add_mean_indicator2(self, g, scatter, bins, probs, pid):
# offset = 0
# percentH = 1 - 0.954 # 2sigma
#
# maxp = max(probs)
# wm, we, mswd, valid_mswd = self._calculate_stats(self.xs, self.xes,
# bins, probs)
# #ym = maxp * percentH + offset
# #set ym in screen space
# #convert to data space
# ogid = self.group_id
# gid = ogid + 1
# sgid = ogid * 3
#
# ym = maxp * 0.1 * gid
#
# s, p = g.new_series(
# [wm], [ym],
# type='scatter',
# marker='circle',
# #selection_marker_size=3,
# marker_size=3,
# #selection_marker='circle',
# #selection_color=scatter.color,
# #selection_outline_color=scatter.color,
# color=scatter.color,
# plotid=0
# )
#
# g.set_series_label('Mean-{}'.format(gid), series=sgid + 2, plotid=pid)
#
# self._add_error_bars(s, [we], 'x', self.options.nsigma)
# # display_mean_indicator = self._get_plot_option(self.options,
# 'display_mean_indicator', default=True)
# if not self.options.display_mean_indicator:
# s.visible = False
#
# label = None
# # display_mean = self._get_plot_option(self.options, 'display_mean_text', default=True)
# if self.options.display_mean:
# text = self._build_label_text(wm, we, mswd, valid_mswd, len(self.xs))
# # font = self._get_plot_option(self.options, 'data_label_font', default='modern 12')
# self._add_data_label(s, text, (wm, ym),
# # font=font
# )
# # add a tool to move the mean age point
# s.tools.append(PointMoveTool(component=s,
# label=label,
# constrain='y'))
def loadData(self,
timer,
condition, # condition as set by another function
bids=[], # if this is set then only load those contigs with these bin ids
verbose=True, # many to some output messages
silent=False, # some to no output messages
loadCovProfiles=True,
loadKmerPCs=True,
loadKmerVarPC=True,
loadRawKmers=False,
makeColors=True,
loadContigNames=True,
loadContigLengths=True,
loadContigGCs=True,
loadBins=False,
loadLinks=False):
"""Load pre-parsed data"""
timer.getTimeStamp()
if(silent):
verbose=False
if verbose:
print "Loading data from:", self.dbFileName
try:
self.numStoits = self.getNumStoits()
self.condition = condition
self.indices = self.dataManager.getConditionalIndices(self.dbFileName,
condition=condition,
silent=silent)
if(verbose):
print " Loaded indices with condition:", condition
self.numContigs = len(self.indices)
if self.numContigs == 0:
print " ERROR: No contigs loaded using condition:", condition
return
if(not silent):
print " Working with: %d contigs" % self.numContigs
if(loadCovProfiles):
if(verbose):
print " Loading coverage profiles"
self.covProfiles = self.dataManager.getCoverageProfiles(self.dbFileName, indices=self.indices)
self.normCoverages = self.dataManager.getNormalisedCoverageProfiles(self.dbFileName, indices=self.indices)
# work out average coverages
self.averageCoverages = np_array([sum(i)/self.numStoits for i in self.covProfiles])
if loadRawKmers:
if(verbose):
print " Loading RAW kmer sigs"
self.kmerSigs = self.dataManager.getKmerSigs(self.dbFileName, indices=self.indices)
if(loadKmerPCs):
self.kmerPCs = self.dataManager.getKmerPCAs(self.dbFileName, indices=self.indices)
if(verbose):
print " Loading PCA kmer sigs (" + str(len(self.kmerPCs[0])) + " dimensional space)"
self.kmerNormPC1 = np_copy(self.kmerPCs[:,0])
self.kmerNormPC1 -= np_min(self.kmerNormPC1)
self.kmerNormPC1 /= np_max(self.kmerNormPC1)
if(loadKmerVarPC):
self.kmerVarPC = self.dataManager.getKmerVarPC(self.dbFileName, indices=self.indices)
if(verbose):
print " Loading PCA kmer variance (total variance: %.2f" % np_sum(self.kmerVarPC) + ")"
if(loadContigNames):
if(verbose):
print " Loading contig names"
self.contigNames = self.dataManager.getContigNames(self.dbFileName, indices=self.indices)
if(loadContigLengths):
self.contigLengths = self.dataManager.getContigLengths(self.dbFileName, indices=self.indices)
if(verbose):
print " Loading contig lengths (Total: %d BP)" % ( sum(self.contigLengths) )
if(loadContigGCs):
self.contigGCs = self.dataManager.getContigGCs(self.dbFileName, indices=self.indices)
if(verbose):
print " Loading contig GC ratios (Average GC: %0.3f)" % ( np_mean(self.contigGCs) )
if(makeColors):
if(verbose):
print " Creating color map"
# use HSV to RGB to generate colors
S = 1 # SAT and VAL remain fixed at 1. Reduce to make
V = 1 # Pastels if that's your preference...
self.colorMapGC = self.createColorMapHSV()
if(loadBins):
if(verbose):
print " Loading bin assignments"
self.binIds = self.dataManager.getBins(self.dbFileName, indices=self.indices)
if len(bids) != 0: # need to make sure we're not restricted in terms of bins
bin_stats = self.getBinStats()
for bid in bids:
try:
self.validBinIds[bid] = bin_stats[bid][0]
self.isLikelyChimeric[bid]= bin_stats[bid][1]
except KeyError:
self.validBinIds[bid] = 0
self.isLikelyChimeric[bid]= False
else:
bin_stats = self.getBinStats()
for bid in bin_stats:
self.validBinIds[bid] = bin_stats[bid][0]
self.isLikelyChimeric[bid] = bin_stats[bid][1]
# fix the binned indices
self.binnedRowIndices = {}
for i in range(len(self.indices)):
if(self.binIds[i] != 0):
self.binnedRowIndices[i] = True
else:
# we need zeros as bin indicies then...
self.binIds = np_zeros(len(self.indices))
if(loadLinks):
self.loadLinks()
self.stoitColNames = self.getStoitColNames()
except:
print "Error loading DB:", self.dbFileName, exc_info()[0]
raise
color = Grid_Values['zGas'].index(parameter_divider)
label = r'$Z = {logage}$'.format(logage = round(float(parameter_divider) * 0.02, 3))
x_values, y_values = Ar_S_model(Line_dict, threshold = 4)
if (x_values != None) and (y_values != None):
dz.data_plot(x_values, y_values, color=dz.ColorVector[2][color], label=label, markerstyle='o')
x_linealFitting = hstack([x_linealFitting, x_values])
y_linealFitting = hstack([y_linealFitting, y_values])
#Lineal model
lineal_mod = LinearModel(prefix='lineal_')
Lineal_parameters = lineal_mod.guess(y_linealFitting, x=x_linealFitting)
x_lineal = linspace(np_min(x_linealFitting), np_max(x_linealFitting), 100)
y_lineal = Lineal_parameters['lineal_slope'].value * x_lineal + Lineal_parameters['lineal_intercept'].value
dz.data_plot(x_lineal, y_lineal, label='Lineal fitting', color = 'black', linestyle='-')
# #Plot fitting formula
formula = r"$log\left(Ar^{{+2}}/Ar^{{+3}}\right) = {m} \cdot log\left(S^{{+2}}/S^{{+3}}\right) + {n}$".format(m=round(Lineal_parameters['lineal_slope'].value,3), n=round(Lineal_parameters['lineal_intercept'].value, 3))
# formula2 = r"$m = {m} \pm {merror}; n = {n} \pm {nerror}$".format(m=round(m[0],3), merror=round(m_err[0],3), n=round(n[0],3), nerror=round(n_err[0],3))
dz.Axis.text(0.50, 0.15, formula, transform=dz.Axis.transAxes, fontsize=20)
# dz.Axis.text(0.50, 0.08, formula2, transform=dz.Axis.transAxes, fontsize=20)
#Plot wording
xtitle = r'$log([SIII]/[SIV])$'
ytitle = r'$log([ArIII]/[ArIV])$'
title = 'Argon - Sulfur ionic relation in Cloudy photoionization models'
dz.FigWording(xtitle, ytitle, title, axis_Size = 20.0, title_Size = 20.0, legend_size=20.0, legend_loc='upper left')
# dz.Axis.set_xlim(0,6)
#Calculate the grid point abundances
#x_values, y_values = Ar_S_model(Line_dict, threshold = 4, z = float(Model_dict['zGas']))
x_values, y_values = Ar_S_abundances_model(Line_dict, diags, Ar3, Ar4, S3, S4, 3)
if (x_values != None) and (y_values != None):
dz.data_plot(x_values, y_values, color=dz.ColorVector[2][color], label=label, markerstyle='o')
x_linealFitting = hstack([x_linealFitting, x_values])
y_linealFitting = hstack([y_linealFitting, y_values])
#Lineal model
lineal_mod = LinearModel(prefix='lineal_')
Lineal_parameters = lineal_mod.guess(y_linealFitting, x=x_linealFitting)
x_lineal = linspace(0, np_max(x_linealFitting), 100)
y_lineal = Lineal_parameters['lineal_slope'].value * x_lineal + Lineal_parameters['lineal_intercept'].value
dz.data_plot(x_lineal, y_lineal, label='Lineal fitting', color = 'black', linestyle='-')
# #Plot fitting formula
formula = r"$log\left(Ar^{{+2}}/Ar^{{+3}}\right) = {m} \cdot log\left(S^{{+2}}/S^{{+3}}\right) + {n}$".format(m=round(Lineal_parameters['lineal_slope'].value,3),
n=round(Lineal_parameters['lineal_intercept'].value, 3))
dz.Axis.text(0.50, 0.15, formula, transform=dz.Axis.transAxes, fontsize=20)
#Plot wording
xtitle = r'$log\left(S^{{+2}}/S^{{+3}}\right)$'
ytitle = r'$log\left(Ar^{{+2}}/Ar^{{+3}}\right)$'
title = 'Argon - Sulfur ionic abundances for several cluster ages and masses'
dz.FigWording(xtitle, ytitle, title, axis_Size = 20.0, title_Size = 20.0, legend_size=20.0, legend_loc='upper left')
#Display figure
def get_value_for_data_only(self, values):
"""
Return the maximum value
"""
return np_max(values)