def plot_part2(filename):
"""
Plots the result of count ones test
"""
fig1 = pl.figure()
iterations, runtimes, fvals = extract(filename)
algos = ["SA", "GA", "MIMIC"]
iters_sa, iters_ga, iters_mimic = [np.array(iterations[a]) for a in algos]
runtime_sa, runtime_ga, runtime_mimic = [np.array(runtimes[a]) for a in algos]
fvals_sa, fvals_ga, fvals_mimic = [np.array(fvals[a]) for a in algos]
plotfunc = getattr(pl, "loglog")
plotfunc(runtime_sa, fvals_sa, "bs", mew=0)
plotfunc(runtime_ga, fvals_ga, "gs", mew=0)
plotfunc(runtime_mimic, fvals_mimic, "rs", mew=0)
# plotfunc(iters_sa, fvals_sa/(runtime_sa * iters_sa), "bs", mew=0)
# plotfunc(iters_ga, fvals_ga/(runtime_ga * iters_ga), "gs", mew=0)
# plotfunc(iters_mimic, fvals_mimic/(runtime_mimic * iters_mimic), "rs", mew=0)
pl.xlabel("Runtime (seconds)")
pl.ylabel("Objective function value")
pl.ylim([min(fvals_sa) / 2, max(fvals_mimic) * 2])
pl.legend(["SA", "GA", "MIMIC"], loc=4)
pl.savefig(filename.replace(".csv", ".png"), bbox_inches="tight")
def plot_field(self,x,y,u=None,v=None,F=None,contour=False,outdir=None,plot='quiver',figname='_field',format='eps'):
outdir = self.set_dir(outdir)
p = 64
if F is None: F=self.calc_F(u,v)
plt.close('all')
plt.figure()
#fig, axes = plt.subplots(nrows=1)
if contour:
plt.hold(True)
plt.contourf(x,y,F)
if plot=='quiver':
plt.quiver(x[::p],y[::p],u[::p],v[::p],scale=0.1)
if plot=='pcolor':
plt.pcolormesh(x[::4],y[::4],F[::4],cmap=plt.cm.Pastel1)
plt.colorbar()
if plot=='stream':
speed = F[::16]
plt.streamplot(x[::16], y[::16], u[::16], v[::16], density=(1,1),color='k')
plt.xlabel('$x$ (a.u.)')
plt.ylabel('$y$ (a.u.)')
plt.savefig(os.path.join(outdir,figname+'.'+format),format=format,dpi=320,bbox_inches='tight')
plt.close()
def plot_feat_hist(data_name_list, filename=None):
if len(data_name_list)>1:
assert filename is not None
pylab.figure(num=None, figsize=(8, 6))
num_rows = 1 + (len(data_name_list) - 1) / 2
num_cols = 1 if len(data_name_list) == 1 else 2
pylab.figure(figsize=(5 * num_cols, 4 * num_rows))
for i in range(num_rows):
for j in range(num_cols):
pylab.subplot(num_rows, num_cols, 1 + i * num_cols + j)
x, name = data_name_list[i * num_cols + j]
pylab.title(name)
pylab.xlabel('Value')
pylab.ylabel('Fraction')
# the histogram of the data
max_val = np.max(x)
if max_val <= 1.0:
bins = 50
elif max_val > 50:
bins = 50
else:
bins = max_val
n, bins, patches = pylab.hist(
x, bins=bins, normed=1, facecolor='blue', alpha=0.75)
pylab.grid(True)
if not filename:
filename = "feat_hist_%s.png" % name.replace(" ", "_")
pylab.savefig(os.path.join(CHART_DIR, filename), bbox_inches="tight")
def check_models(self):
plt.figure('Bandgap narrowing')
Na = np.logspace(12, 20)
Nd = 0.
dn = 1e14
temp = 300.
for author in self.available_models():
BGN = self.update(Na=Na, Nd=Nd, nxc=dn,
author=author,
temp=temp)
if not np.all(BGN == 0):
plt.plot(Na, BGN, label=author)
test_file = os.path.join(
os.path.dirname(os.path.realpath(__file__)),
'Si', 'check data', 'Bgn.csv')
data = np.genfromtxt(test_file, delimiter=',', names=True)
for name in data.dtype.names[1:]:
plt.plot(
data['N'], data[name], 'r--',
label='PV-lighthouse\'s: ' + name)
plt.semilogx()
plt.xlabel('Doping (cm$^{-3}$)')
plt.ylabel('Bandgap narrowing (K)')
plt.legend(loc=0)
def plot_bernoulli_matrix(self, show_npfs=False):
"""
Plot the heatmap of the Bernoulli matrix
@self
@show_npfs - Highlight NPFS detections [Boolean]
"""
matrix = self.Bernoulli_matrix
if show_npfs == False:
plot = plt.imshow(matrix)
plot.set_cmap('hot')
plt.colorbar()
plt.xlabel("Bootstraps")
plt.ylabel("Feature")
plt.show()
else:
for i in self.selected_features:
for k in range(len(matrix[i])):
matrix[i,k] = .5
plot = plt.imshow(matrix)
plot.set_cmap('hot')
plt.xlabel("Bootstraps")
plt.ylabel("Feature")
plt.colorbar()
plt.show()
return None
def study_sdss_density(hemi='south'):
grid = grid3d(hemi=hemi)
n_data = num_sdss_data_both_catalogs(hemi, grid)
n_rand, weight = num_sdss_rand_both_catalogs(hemi, grid)
n_rand *= ((n_data*weight).sum() / (n_rand*weight).sum())
delta = (n_data - n_rand) / n_rand
delta[weight==0]=0.
fdelta = np.fft.fftn(delta*weight)
power = np.abs(fdelta)**2.
ks = get_wavenumbers(delta.shape, grid.reso_mpc)
kmag = ks[3]
kbin = np.arange(0,0.06,0.002)
ind = np.digitize(kmag.ravel(), kbin)
power_ravel = power.ravel()
power_bin = np.zeros_like(kbin)
for i in range(len(kbin)):
print i
wh = np.where(ind==i)[0]
power_bin[i] = power_ravel[wh].mean()
#pl.clf()
#pl.plot(kbin, power_bin)
from cosmolopy import perturbation
pk = perturbation.power_spectrum(kbin, 0.4, **cosmo)
pl.clf(); pl.plot(kbin, power_bin/pk, 'b')
pl.plot(kbin, power_bin/pk, 'bo')
pl.xlabel('k (1/Mpc)',fontsize=16)
pl.ylabel('P(k) ratio, DATA/THEORY [arb. norm.]',fontsize=16)
ipdb.set_trace()
def make_corr1d_fig(dosave=False):
corr = make_corr_both_hemi()
lw=2; fs=16
pl.figure(1)#, figsize=(8, 7))
pl.clf()
pl.xlim(4,300)
pl.ylim(-400,+500)
lambda_titles = [r'$20 < \lambda < 30$',
r'$30 < \lambda < 40$',
r'$\lambda > 40$']
colors = ['blue','green','red']
for i in range(3):
corr1d, rcen = corr_1d_from_2d(corr[i])
ipdb.set_trace()
pl.semilogx(rcen, corr1d*rcen**2, lw=lw, color=colors[i])
#pl.semilogx(rcen, corr1d*rcen**2, 'o', lw=lw, color=colors[i])
pl.xlabel(r'$s (Mpc)$',fontsize=fs)
pl.ylabel(r'$s^2 \xi_0(s)$', fontsize=fs)
pl.legend(lambda_titles, 'lower left', fontsize=fs+3)
pl.plot([.1,10000],[0,0],'k--')
s_bao = 149.28
pl.plot([s_bao, s_bao],[-9e9,+9e9],'k--')
pl.text(s_bao*1.03, 420, 'BAO scale')
pl.text(s_bao*1.03, 370, '%0.1f Mpc'%s_bao)
if dosave: pl.savefig('xi1d_3bin.pdf')
def study_redmapper_2d():
# I just want to know the typical angular separation for RM clusters.
# I'm going to do this in a lazy way.
hemi = 'north'
rm = load_redmapper(hemi=hemi)
ra = rm['ra']
dec = rm['dec']
ncl = len(ra)
dist = np.zeros((ncl, ncl))
for i in range(ncl):
this_ra = ra[i]
this_dec = dec[i]
dra = this_ra-ra
ddec = this_dec-dec
dxdec = dra*np.cos(this_dec*np.pi/180.)
dd = np.sqrt(dxdec**2. + ddec**2.)
dist[i,:] = dd
dist[i,i] = 99999999.
d_near_arcmin = dist.min(0)*60.
pl.clf(); pl.hist(d_near_arcmin, bins=100)
pl.title('Distance to Nearest Neighbor for RM clusters')
pl.xlabel('Distance (arcmin)')
pl.ylabel('N')
fwhm_planck_217 = 5.5 # arcmin
sigma = fwhm_planck_217/2.355
frac_2sigma = 1.*len(np.where(d_near_arcmin>2.*sigma)[0])/len(d_near_arcmin)
frac_3sigma = 1.*len(np.where(d_near_arcmin>3.*sigma)[0])/len(d_near_arcmin)
print '%0.3f percent of RM clusters are separated by 2-sigma_planck_beam'%(100.*frac_2sigma)
print '%0.3f percent of RM clusters are separated by 3-sigma_planck_beam'%(100.*frac_3sigma)
ipdb.set_trace()
def study_multiband_planck(quick=True):
savename = datadir+'cl_multiband.pkl'
bands = [100, 143, 217, 'mb']
if quick: cl = pickle.load(open(savename,'r'))
else:
cl = {}
mask = load_planck_mask()
mask_factor = np.mean(mask**2.)
for band in bands:
this_map = load_planck_data(band)
this_cl = hp.anafast(this_map*mask, lmax=lmax)/mask_factor
cl[band] = this_cl
pickle.dump(cl, open(savename,'w'))
cl_theory = {}
pl.clf()
for band in bands:
l_theory, cl_theory[band] = get_cl_theory(band)
this_cl = cl[band]
pl.plot(this_cl/cl_theory[band])
pl.legend(bands)
pl.plot([0,4000],[1,1],'k--')
pl.ylim(.7,1.3)
pl.ylabel('data/theory')
def plot(self, title=None, **kwargs):
"""Generates a pylab plot from the result set.
``matplotlib`` must be installed, and in an
IPython Notebook, inlining must be on::
%%matplotlib inline
The first and last columns are taken as the X and Y
values. Any columns between are ignored.
Parameters
----------
title: Plot title, defaults to names of Y value columns
Any additional keyword arguments will be passsed
through to ``matplotlib.pylab.plot``.
"""
import matplotlib.pylab as plt
self.guess_plot_columns()
self.x = self.x or range(len(self.ys[0]))
coords = reduce(operator.add, [(self.x, y) for y in self.ys])
plot = plt.plot(*coords, **kwargs)
if hasattr(self.x, 'name'):
plt.xlabel(self.x.name)
ylabel = ", ".join(y.name for y in self.ys)
plt.title(title or ylabel)
plt.ylabel(ylabel)
return plot
def bar(self, key_word_sep = " ", title=None, **kwargs):
"""Generates a pylab bar plot from the result set.
``matplotlib`` must be installed, and in an
IPython Notebook, inlining must be on::
%%matplotlib inline
The last quantitative column is taken as the Y values;
all other columns are combined to label the X axis.
Parameters
----------
title: Plot title, defaults to names of Y value columns
key_word_sep: string used to separate column values
from each other in labels
Any additional keyword arguments will be passsed
through to ``matplotlib.pylab.bar``.
"""
import matplotlib.pylab as plt
self.guess_pie_columns(xlabel_sep=key_word_sep)
plot = plt.bar(range(len(self.ys[0])), self.ys[0], **kwargs)
if self.xlabels:
plt.xticks(range(len(self.xlabels)), self.xlabels,
rotation=45)
plt.xlabel(self.xlabel)
plt.ylabel(self.ys[0].name)
return plot
def test_simple_gen(self):
self_con = .8
other_con = 0.05
g = self.gen.gen_stoch_blockmodel(min_degree=1, blocks=5, self_con=self_con, other_con=other_con,
powerlaw_exp=2.1, degree_seq='powerlaw', num_nodes=1000, num_links=3000)
deg_hist = vertex_hist(g, 'total')
res = fit_powerlaw.Fit(g.degree_property_map('total').a, discrete=True)
print 'powerlaw alpha:', res.power_law.alpha
print 'powerlaw xmin:', res.power_law.xmin
if len(deg_hist[0]) != len(deg_hist[1]):
deg_hist[1] = deg_hist[1][:len(deg_hist[0])]
print 'plot degree dist'
plt.plot(deg_hist[1], deg_hist[0])
plt.xscale('log')
plt.xlabel('degree')
plt.ylabel('#nodes')
plt.yscale('log')
plt.savefig('deg_dist_test.png')
plt.close('all')
print 'plot graph'
pos = sfdp_layout(g, groups=g.vp['com'], mu=3)
graph_draw(g, pos=pos, output='graph.png', output_size=(800, 800),
vertex_size=prop_to_size(g.degree_property_map('total'), mi=2, ma=30), vertex_color=[0., 0., 0., 1.],
vertex_fill_color=g.vp['com'],
bg_color=[1., 1., 1., 1.])
plt.close('all')
print 'init:', self_con / (self_con + other_con), other_con / (self_con + other_con)
print 'real:', gt_tools.get_graph_com_connectivity(g, 'com')
def check_models(self):
'''
Displays a plot of the models against that taken from a
respected website (https://www.pvlighthouse.com.au/)
'''
plt.figure('Intrinsic bandgap')
t = np.linspace(1, 500)
for author in self.available_models():
Eg = self.update(temp=t, author=author, multiplier=1.0)
plt.plot(t, Eg, label=author)
test_file = os.path.join(
os.path.dirname(os.path.realpath(__file__)),
'Si', 'check data', 'iBg.csv')
data = np.genfromtxt(test_file, delimiter=',', names=True)
for temp, name in zip(data.dtype.names[0::2], data.dtype.names[1::2]):
plt.plot(
data[temp], data[name], '--', label=name)
plt.xlabel('Temperature (K)')
plt.ylabel('Intrinsic Bandgap (eV)')
plt.legend(loc=0)
self.update(temp=0, author=author, multiplier=1.01)
def plot_confusion_matrix(cm, title='', cmap=plt.cm.Blues):
#print cm
#display vehicle, idle, walking accuracy respectively
#display overall accuracy
print type(cm)
# plt.figure(index
plt.imshow(cm, interpolation='nearest', cmap=cmap)
#plt.figure("")
plt.title("Confusion Matrix")
plt.colorbar()
tick_marks = [0,1,2]
target_name = ["driving","idling","walking"]
plt.xticks(tick_marks,target_name,rotation=45)
plt.yticks(tick_marks,target_name,rotation=45)
print len(cm[0])
for i in range(0,3):
for j in range(0,3):
plt.text(i,j,str(cm[i,j]))
plt.tight_layout()
plt.ylabel("Actual Value")
plt.xlabel("Predicted Outcome")
def plot_values(self, TITLE, SAVE):
plot(self.list_of_densities, self.list_of_pressures)
title(TITLE)
xlabel("Densities")
ylabel("Pressure")
savefig(SAVE)
show()
def __call__(self, **params):
p = ParamOverrides(self, params)
fig = plt.figure(figsize=(5, 5))
# This one-liner works in Octave, but in matplotlib it
# results in lines that are all connected across rows and columns,
# so here we plot each line separately:
# plt.plot(x,y,"k-",transpose(x),transpose(y),"k-")
# Here, the "k-" means plot in black using solid lines;
# see matplotlib for more info.
isint = plt.isinteractive() # Temporarily make non-interactive for
# plotting
plt.ioff()
for r, c in zip(p.y[::p.skip], p.x[::p.skip]):
plt.plot(c, r, "k-")
for r, c in zip(np.transpose(p.y)[::p.skip],np.transpose(p.x)[::p.skip]):
plt.plot(c, r, "k-")
# Force last line avoid leaving cells open
if p.skip != 1:
plt.plot(p.x[-1], p.y[-1], "k-")
plt.plot(np.transpose(p.x)[-1], np.transpose(p.y)[-1], "k-")
plt.xlabel('x')
plt.ylabel('y')
# Currently sets the input range arbitrarily; should presumably figure out
# what the actual possible range is for this simulation (which would presumably
# be the maximum size of any GeneratorSheet?).
plt.axis(p.axis)
if isint: plt.ion()
self._generate_figure(p)
return fig
def inter_show(start, lc, eta, vol_ins, props, lbl_outs, grdts, pars):
'''
Plots a display of training information to the screen
'''
import matplotlib.pylab as plt
name_in, vol = vol_ins.popitem()
name_p, prop = props.popitem()
name_l, lbl = lbl_outs.popitem()
name_g, grdt = grdts.popitem()
m_input = volume_util.crop(vol[0,:,:,:], prop.shape[-3:]) #good enough for now
# real time visualization
plt.subplot(251), plt.imshow(vol[0,0,:,:], interpolation='nearest', cmap='gray')
plt.xlabel('input')
plt.subplot(252), plt.imshow(m_input[0,:,:], interpolation='nearest', cmap='gray')
plt.xlabel('matched input')
plt.subplot(253), plt.imshow(prop[0,0,:,:], interpolation='nearest', cmap='gray')
plt.xlabel('output')
plt.subplot(254), plt.imshow(lbl[0,0,:,:], interpolation='nearest', cmap='gray')
plt.xlabel('label')
plt.subplot(255), plt.imshow(grdt[0,0,:,:], interpolation='nearest', cmap='gray')
plt.xlabel('gradient')
plt.subplot(256)
plt.plot(lc.tn_it, lc.tn_err, 'b', label='train')
plt.plot(lc.tt_it, lc.tt_err, 'r', label='test')
plt.xlabel('iteration'), plt.ylabel('cost energy')
plt.subplot(257)
plt.plot( lc.tn_it, lc.tn_cls, 'b', lc.tt_it, lc.tt_cls, 'r')
plt.xlabel('iteration'), plt.ylabel( 'classification error' )
return
def plot_feat_hist(data_name_list, filename=None):
pylab.clf()
# import pdb;pdb.set_trace()
num_rows = 1 + (len(data_name_list) - 1) / 2
num_cols = 1 if len(data_name_list) == 1 else 2
pylab.figure(figsize=(5 * num_cols, 4 * num_rows))
for i in range(num_rows):
for j in range(num_cols):
pylab.subplot(num_rows, num_cols, 1 + i * num_cols + j)
x, name = data_name_list[i * num_cols + j]
pylab.title(name)
pylab.xlabel('Value')
pylab.ylabel('Density')
# the histogram of the data
max_val = np.max(x)
if max_val <= 1.0:
bins = 50
elif max_val > 50:
bins = 50
else:
bins = max_val
n, bins, patches = pylab.hist(
x, bins=bins, normed=1, facecolor='green', alpha=0.75)
pylab.grid(True)
if not filename:
filename = "feat_hist_%s.png" % name
pylab.savefig(os.path.join(CHART_DIR, filename), bbox_inches="tight")
def cdf(x,colsym="",lab="",lw=4):
""" plot the cumulative density function
Parameters
----------
x : np.array()
colsym : string
lab : string
lw : int
linewidth
Examples
--------
>>> import numpy as np
"""
rcParams['legend.fontsize']=20
rcParams['font.size']=20
x = np.sort(x)
n = len(x)
x2 = np.repeat(x, 2)
y2 = np.hstack([0.0, repeat(np.arange(1,n) / float(n), 2), 1.0])
plt.plot(x2,y2,colsym,label=lab,linewidth=lw)
plt.grid('on')
plt.legend(loc=2)
plt.xlabel('Ranging Error[m]')
plt.ylabel('Cumulative Probability')
def plot_peaks(spectra, ref, zl, pl, filename=''):
plt.figure();
plt.title("Debug: Peak fitting for '%s'" % filename);
plt.xlabel("y-position [px]");
plt.ylabel("Intensity");
Nspectra, Npx = spectra.shape;
for s in xrange(Nspectra):
scale = 1./spectra.max();
offset= -s*0.1;
# plot data
plt.plot(spectra[s]*scale + offset,'k',linewidth=2);
# plot first peak
p,A,w = zl[s];
x = np.arange(-2*w, 2*w) + p;
plt.plot(x,gauss(x,*zl[s])*scale + offset,'r');
# plot second peak
if ref is not None:
p,A = pl[s];
x = np.arange(len(ref)) - len(ref)/2 + p;
plt.plot(x,ref/ref.max()*A*scale + offset,'g');
def fancy_dendrogram(*args, **kwargs):
'''
Source: https://joernhees.de/blog/2015/08/26/scipy-hierarchical-clustering-and-dendrogram-tutorial/
'''
from scipy.cluster import hierarchy
import matplotlib.pylab as plt
max_d = kwargs.pop('max_d', None)
if max_d and 'color_threshold' not in kwargs:
kwargs['color_threshold'] = max_d
annotate_above = kwargs.pop('annotate_above', 0)
ddata = hierarchy.dendrogram(*args, **kwargs)
if not kwargs.get('no_plot', False):
plt.title('Hierarchical Clustering Dendrogram (truncated)')
plt.xlabel('sample index or (cluster size)')
plt.ylabel('distance')
for i, d, c in zip(ddata['icoord'], ddata['dcoord'], ddata['color_list']):
x = 0.5 * sum(i[1:3])
y = d[1]
if y > annotate_above:
plt.plot(x, y, 'o', c=c)
plt.annotate("%.3g" % y, (x, y), xytext=(0, -5),
textcoords='offset points',
va='top', ha='center')
if max_d:
plt.axhline(y=max_d, c='k')
return ddata
def modelfit(alg, train, target, test, useTrainCV=True, cv_folds=5, early_stopping_rounds=50):
if useTrainCV:
xgboost_params = alg.get_xgb_params()
xgtrain = xgb.DMatrix(train.values, label=target.values)
xgtest = xgb.DMatrix(test.values)
watchlist = [(xgtrain, 'train')] # Specify validations set to watch performance
cvresult = xgb.cv(xgboost_params, xgtrain, num_boost_round=alg.get_params()['n_estimators'], nfold=cv_folds, early_stopping_rounds=early_stopping_rounds) #metrics='auc',show_progress=False
alg.set_params(n_estimators=cvresult.shape[0])
# Fit the algorithm on the data
alg.fit(train, target, eval_metric='auc')
# Predict training set:
train_preds = alg.predict(train)
train_predprob = alg.predict_proba(train)[:,1]
# Print model report:
print "\nModel Report"
print "Accuracy : %.4g" % metrics.accuracy_score(target.values, train_preds)
print "AUC Score (Train): %f" % metrics.roc_auc_score(target, train_predprob)
# Make a prediction:
print('Predicting......')
test_predprob = alg.predict_proba(test)[:,1]
feat_imp = pd.Series(alg.booster().get_fscore()).sort_values(ascending=False)
feat_imp.plot(kind='bar', title='Feature Importances')
plt.ylabel('Feature Importance Score')
#plt.show()
return test_predprob
def fdr(p_values=None, verbose=0):
"""Returns the FDR associated with each p value
Parameters
-----------
p_values : ndarray of shape (n)
The samples p-value
Returns
-------
q : array of shape(n)
The corresponding fdr values
"""
p_values = check_p_values(p_values)
n_samples = p_values.size
order = p_values.argsort()
sp_values = p_values[order]
# compute q while in ascending order
q = np.minimum(1, n_samples * sp_values / np.arange(1, n_samples + 1))
for i in range(n_samples - 1, 0, - 1):
q[i - 1] = min(q[i], q[i - 1])
# reorder the results
inverse_order = np.arange(n_samples)
inverse_order[order] = np.arange(n_samples)
q = q[inverse_order]
if verbose:
import matplotlib.pylab as mp
mp.figure()
mp.xlabel('Input p-value')
mp.plot(p_values, q, '.')
mp.ylabel('Associated fdr')
return q
def plot_q(model='cem', r_min=0.0, r_max=6371.0, dr=1.0):
"""
Plot a radiallysymmetric Q model.
plot_q(model='cem', r_min=0.0, r_max=6371.0, dr=1.0):
r_min=minimum radius [km], r_max=maximum radius [km], dr=radius
increment [km]
Currently available models (model): cem, prem, ql6
"""
import matplotlib.pylab as plt
r = np.arange(r_min, r_max + dr, dr)
q = np.zeros(len(r))
for k in range(len(r)):
if model == 'cem':
q[k] = q_cem(r[k])
elif model == 'ql6':
q[k] = q_ql6(r[k])
elif model == 'prem':
q[k] = q_prem(r[k])
plt.plot(r, q, 'k')
plt.xlim((0.0, r_max))
plt.xlabel('radius [km]')
plt.ylabel('Q')
plt.show()
def handle(self, *args, **options):
try:
from matplotlib import pylab as pl
import numpy as np
except ImportError:
raise Exception('Be sure to install requirements_scipy.txt before running this.')
all_names_and_counts = RawCommitteeTransactions.objects.all().values('attest_by_name').annotate(total=Count('attest_by_name')).order_by('-total')
all_names_and_counts_as_tuple_and_sorted = sorted([(row['attest_by_name'], row['total']) for row in all_names_and_counts], key=lambda row: row[1])
print "top ten attestors: (name, number of transactions they attest for)"
for row in all_names_and_counts_as_tuple_and_sorted[-10:]:
print row
n_bins = 100
filename = 'attestor_participation_distribution.png'
x_max = all_names_and_counts_as_tuple_and_sorted[-31][1] # eliminate top outliers from hist
x_min = all_names_and_counts_as_tuple_and_sorted[0][1]
counts = [row['total'] for row in all_names_and_counts]
pl.figure(1, figsize=(18, 6))
pl.hist(counts, bins=np.arange(x_min, x_max, (float(x_max)-x_min)/100) )
pl.title('Histogram of Attestor Participation in RawCommitteeTransactions')
pl.xlabel('Number of transactions a person attested for')
pl.ylabel('Number of people')
pl.savefig(filename)
def plot_runtime_results(results):
plt.rcParams["figure.figsize"] = 7,7
plt.rcParams["font.size"] = 22
matplotlib.rc("xtick", labelsize=24)
matplotlib.rc("ytick", labelsize=24)
params = {"text.fontsize" : 32,
"font.size" : 32,
"legend.fontsize" : 30,
"axes.labelsize" : 32,
"text.usetex" : False
}
plt.rcParams.update(params)
#plt.semilogx(results[:,0], results[:,3], 'r-x', lw=3)
#plt.semilogx(results[:,0], results[:,1], 'g-D', lw=3)
#plt.semilogx(results[:,0], results[:,2], 'b-s', lw=3)
plt.plot(results[:,0], results[:,3], 'r-x', lw=3, ms=10)
plt.plot(results[:,0], results[:,1], 'g-D', lw=3, ms=10)
plt.plot(results[:,0], results[:,2], 'b-s', lw=3, ms=10)
plt.legend(["Chain", "Tree", "FFT Tree"], loc="upper left")
plt.xticks([1e5, 2e5, 3e5])
plt.yticks([0, 60, 120, 180])
plt.xlabel("Problem Size")
plt.ylabel("Runtime (sec)")
return results
def flipPlot(minExp, maxExp):
"""假定minEXPy和maxExp是正整数且minExp<maxExp
绘制出2**minExp到2**maxExp次抛硬币的结果
"""
ratios = []
diffs = []
aAxis = []
for i in range(minExp, maxExp+1):
aAxis.append(2**i)
for numFlips in aAxis:
numHeads = 0
for n in range(numFlips):
if random.random() < 0.5:
numHeads += 1
numTails = numFlips - numHeads
ratios.append(numHeads/numFlips)
diffs.append(abs(numHeads-numTails))
plt.figure()
ax1 = plt.subplot(121)
plt.title("Difference Between Heads and Tails")
plt.xlabel('Number of Flips')
plt.ylabel('Abs(#Heads - #Tails)')
ax1.semilogx(aAxis, diffs, 'bo')
ax2 = plt.subplot(122)
plt.title("Heads/Tails Ratios")
plt.xlabel('Number of Flips')
plt.ylabel("#Heads/#Tails")
ax2.semilogx(aAxis, ratios, 'bo')
plt.show()
def plot_corner_posteriors(self, savefile=None, labels=["T1", "R1", "Av", "T2", "R2"]):
'''
Plots the corner plot of the MCMC results.
'''
ndim = len(self.sampler.flatchain[0,:])
chain = self.sampler
samples = chain.flatchain
samples = samples[:,0:ndim]
plt.figure(figsize=(8,8))
fig = corner.corner(samples, labels=labels[0:ndim])
plt.title("MJD: %.2f"%self.mjd)
name = self._get_save_path(savefile, "mcmc_posteriors")
plt.savefig(name)
plt.close("all")
plt.figure(figsize=(8,ndim*3))
for n in range(ndim):
plt.subplot(ndim,1,n+1)
chain = self.sampler.chain[:,:,n]
nwalk, nit = chain.shape
for i in np.arange(nwalk):
plt.plot(chain[i], lw=0.1)
plt.ylabel(labels[n])
plt.xlabel("Iteration")
name_walkers = self._get_save_path(savefile, "mcmc_walkers")
plt.tight_layout()
plt.savefig(name_walkers)
plt.close("all")
def plotMassFunction(im, pm, outbase, mmin=9, mmax=13, mstep=0.05):
"""
Make a comparison plot between the input mass function and the
predicted projected correlation function
"""
plt.clf()
nmbins = ( mmax - mmin ) / mstep
mbins = np.logspace( mmin, mmax, nmbins )
mcen = ( mbins[:-1] + mbins[1:] ) /2
plt.xscale( 'log', nonposx = 'clip' )
plt.yscale( 'log', nonposy = 'clip' )
ic, e, p = plt.hist( im, mbins, label='Original Halos', alpha=0.5, normed = True)
pc, e, p = plt.hist( pm, mbins, label='Added Halos', alpha=0.5, normed = True)
plt.legend()
plt.xlabel( r'$M_{vir}$' )
plt.ylabel( r'$\frac{dN}{dM}$' )
#plt.tight_layout()
plt.savefig( outbase+'_mfcn.png' )
mdtype = np.dtype( [ ('mcen', float), ('imcounts', float), ('pmcounts', float) ] )
mf = np.ndarray( len(mcen), dtype = mdtype )
mf[ 'mcen' ] = mcen
mf[ 'imcounts' ] = ic
mf[ 'pmcounts' ] = pc
fitsio.write( outbase+'_mfcn.fit', mf )
def plot_cell(self,cell_number=0,label='insert_label'):
current_cell = self.cell_list[cell_number]
temp = current_cell.temp
cd_signal = current_cell.cd_signal
cd_calc = current_cell.cd_calc()
ax = pylab.gca()
pylab.plot(temp,cd_signal,'o',color='black')
pylab.plot(temp,cd_calc,color='black')
pylab.xlabel(r'Temperature ($^{\circ}$C)')
pylab.ylabel('mdeg')
pylab.ylim([-25,-4])
dH = numpy.round(current_cell.dH, decimals=1)
Tm = numpy.round(current_cell.Tm-273.15, decimals=1)
nf = current_cell.nf
nu = current_cell.nu
textstr_dH = '${\Delta}H_{m}$ = %.1f kcal/mol' %dH
textstr_Tm ='$T_{m}$ = %.1f $^{\circ}$C' %Tm
textstr_nf ='$N_{folded}$ = %d' %nf
textstr_nu ='$N_{unfolded}$ = %d'%nu
ax.text(8,-6,textstr_dH, fontsize=16,ha='left',va='top')
ax.text(8,-7.5,textstr_Tm, fontsize=16,ha='left',va='top')
ax.text(8,-9,textstr_nf, fontsize=16,ha='left',va='top')
ax.text(8,-10.5,textstr_nu, fontsize=16,ha='left',va='top')
pylab.title(label)
pylab.show()
return
#%%
# The fourth subplot, which shows the optimization of Cs
def make_patch_spines_invisible(ax):
ax.set_frame_on(True)
ax.patch.set_visible(False)
for sp in ax.spines.values():
sp.set_visible(False)
dsWD = [4, 7, 4, 7] # the line dash of the first configuration
ax = plt.subplot(grid[1, 13:]) #plt.subplot(2,4,7)
plt.title('(d)', fontsize=fs)
plt.xlabel('$c_s$ ', fontsize=fs)
plt.ylabel('Mean $W_d$ (10$^{5}$ km)', fontsize=fs, color=color1)
# the wasserstein axis
#ax.set_yticklabels(labels=[-2,-1,0,1,2,3,4,5],fontsize=fs, color='k')
#a0 = sns.lineplot(x=CS,y=lsq/10**5, linewidth = lw, ax=ax, color=color1,
# zorder=9)
a0 = sns.lineplot(x=CS,
y=np.full(len(lsq), mean_distance_mm / 10**5),
linewidth=lw,
ax=ax,
color=color1,
zorder=10)
a0.lines[0].set_linestyle(":")
sns.lineplot(x=CS,
def train(self,
X_train,
Y_train,
X_test,
Y_test,
print_accuracy=False,
show_loss_graph=False,
show_accuracy_graph=False):
x_train, y_train = self.normalize_vec(X_train, Y_train)
x_test, y_test = self.normalize_vec(X_test, Y_test)
train_loss = []
train_acc_histroy = []
test_acc_history = []
params = {}
for i in range(self.epoch_num):
loss_per_epoch = 0
accuracy_per_epoch = 0
minibatches = shuffle_batches(x_train, y_train, self.batch_size)
for j in range(len(minibatches)):
x_batch = minibatches[j][0]
y_batch = minibatches[j][1]
grads = self.gradient(x_batch, y_batch)
params = {
'W1': self.W1,
'b1': self.b1,
'W2': self.W2,
'b2': self.b2,
'gamma': self.gamma,
'beta': self.beta
}
self.optimizer.update(params, grads)
if self.use_bn:
self.running_mean = self.layers['BatchNorm'].running_mean
self.running_var = self.layers['BatchNorm'].running_var
loss = self.loss(x_batch, y_batch)
loss_per_epoch += loss
ave_loss = loss_per_epoch / len(minibatches)
train_loss.append(ave_loss)
if print_accuracy:
train_accuracy = self.accuracy(x_train, y_train)
test_accuracy = self.accuracy(x_test, y_test)
train_acc_histroy.append(train_accuracy / X_train.shape[0])
test_acc_history.append(test_accuracy / X_test.shape[0])
print("Epoch {i}: Loss={ave_loss} Accuracy_train={train_acc:.4f} Accuracy_test={test_acc:.4f}"\
.format(i=i,ave_loss=ave_loss,train_acc=train_accuracy/X_train.shape[0],test_acc=test_accuracy/X_test.shape[0]))
else:
print("Epoch {i}: Loss={ave_loss}".format(i))
print("Final test_accuracy: {acc}".format(
acc=self.accuracy(x_test, y_test) / X_test.shape[0]))
path = os.path.dirname(os.path.abspath(__file__))
if show_loss_graph:
x = np.linspace(0, self.epoch_num, self.epoch_num)
plt.plot(x, train_loss, label='loss')
plt.title('Average Loss of Each Epoch--{activation}'.format(
activation=self.activation))
plt.xlabel('epoch')
plt.ylabel('loss')
plt.xlim(left=0)
plt.ylim(bottom=0)
# plt.savefig(os.path.join(path,'fig_loss_{act_f}_{mid_node_num}n_{epoch_num}e.png'\
# .format(act_f=self.activation,mid_node_num=self.sizes[1], epoch_num=self.epoch_num)))
plt.show()
if show_accuracy_graph:
x2 = np.linspace(0, len(test_acc_history), len(test_acc_history))
plt.plot(x2, train_acc_histroy, label='train accuracy')
plt.plot(x2,
test_acc_history,
label='test accuracy',
linestyle='--')
plt.title('Accuracy After Each Epoch--{activation}'.format(
activation=self.activation))
plt.xlabel('epoch')
plt.ylabel('accuracy')
plt.xlim(left=0)
plt.ylim(0, 1.0)
plt.legend(loc='lower right')
# plt.savefig(os.path.join(path,'fig_acc_{act_f}_{mid_node_num}n_{epoch_num}e.png'\
# .format(act_f=self.activation,mid_node_num=self.sizes[1], epoch_num=self.epoch_num)))
plt.show()
if self.use_bn:
params['W1'], params['b1'], params['W2'], params['b2'] ,params['gamma'], params['beta'], params['running_mean'], params['running_var']= \
self.W1, self.b1, self.W2, self.b2, self.gamma, self.beta, self.running_mean,self.running_var
else:
params['W1'], params['b1'], params['W2'], params['b2'] ,params['gamma'], params['beta']= \
self.W1, self.b1, self.W2, self.b2, self.gamma, self.beta
print("Network Architecture:{a_f}-Dropout({use_dropout}({drop_p}))-BatchNorm({use_bn})"\
.format(a_f=self.activation,use_dropout=self.use_dropout,drop_p=self.dropout_p, use_bn=self.use_bn))
print("Training, Done!")
return params
return a * b
h = np.linspace(0.0001, 1 - 1e-5, 100)
y = prob(h, cadena)
m = np.trapz(y, x=h)
y_new = y / m
#Valores importantes
a = np.where(y_new == np.max(y_new))
H_0 = h[a[0][0]]
valor = np.log(y_new)[a[0][0] +
1] - 2 * np.log(y_new)[a[0][0]] + np.log(y_new)[a[0][0] -
1]
deriv2 = valor / (H_0 - h[a[0][0] - 1])**2
sigma = 1 / np.sqrt(-deriv2)
def aprox(h):
A = 1.0 / (sigma * (np.sqrt(2 * np.pi)))
B = np.exp(-(h - H_0)**2 / (2 * (sigma**2)))
return A * B
plt.plot(h, aprox(h), '--')
plt.plot(h, y_new)
plt.xlabel('H')
plt.ylabel('P(H|{datos})')
plt.title('H = {:.2f} {} {:.2f}'.format(H_0, '$\pm$', sigma))
plt.savefig('coins.png')
for i in k:
kmeans = KMeans(n_clusters=i)
kmeans.fit(df_norm)
WSS = [] # variable for storing within sum of squares for each cluster
for j in range(i):
WSS.append(
sum(
cdist(df_norm.iloc[kmeans.labels_ == j, :],
kmeans.cluster_centers_[j].reshape(1, df_norm.shape[1]),
"euclidean")))
TWSS.append(sum(WSS))
# Scree plot
plt.plot(k, TWSS, 'ro-')
plt.xlabel("No_of_Clusters")
plt.ylabel("total_within_SS")
plt.xticks(k)
# Selecting 5 clusters from the above scree plot which is the optimum number of clusters
model = KMeans(n_clusters=5)
model.fit(df_norm)
model.labels_ # getting the labels of clusters assigned to each row
md = pd.Series(
model.labels_) # converting numpy array into pandas series object
Univ['clust'] = md # creating a new column and assigning it to new column
df_norm.head()
Univ = Univ.iloc[:, [7, 0, 1, 2, 3, 4, 5, 6]]
Univ.iloc[:, 1:7].groupby(Univ.clust).mean()
0,
self.window + 2,
color="k",
linestyles="dashed")
plt.xlabel("Lag")
plt.ylabel("Autocorrelation")
plt.title("Correlogram for Forecast Error")
plt.show()
hw = HoltWinters("demand.csv", window=12)
hw.fit_holt_winters()
forecast_one_step, forecast_error = hw.holt_winters_best_fit()
future_mean, future_lb, future_ub = hw.predict(forecast_n_steps=12, n_sims=500)
# Plot correlogram
hw.plot_correlogram(forecast_error)
# Plot forecast
plt.plot(hw.ts.keys(), hw.ts.values(), "-k", label="raw")
plt.plot(hw.ts.keys(), forecast_one_step, "--k", label="holt-winters")
x_future_min = max(hw.ts.keys()) + 1
x_future = range(x_future_min, x_future_min + len(future_mean))
plt.plot(x_future, future_mean, color='green', label="95% CI")
plt.fill_between(x_future, future_ub, future_lb, color="green", alpha=0.3)
plt.legend(loc="best")
plt.xlabel("Date Index")
plt.ylabel("Metric")
plt.show()
trainData = pca.transform(trainData)
testData = pca.transform(testData)
totalData = numpy.concatenate((trainData, testData), axis=0)
totalData = scale(totalData)
trainData = totalData[0:len(trainData)]
testData = totalData[len(trainData):]
# print(trainData[0:5])
clf = SVC(probability=True)
clf.fit(trainData, trainLabel)
# joblib.dump(clf, savepath + 'Parameter.m')
probability = clf.predict_proba(testData)
fpr, tpr, thresholds = roc_curve(testLabel, probability[:, 1])
mean_tpr += interp(mean_fpr, fpr,
tpr) # 对mean_tpr在mean_fpr处进行插值，通过scipy包调用interp()函数
mean_tpr[0] = 0.0 # 初始处为0
roc_auc = auc(fpr, tpr)
plt.plot(fpr, tpr, lw=1, label='ROC fold %d' % appoint)
aucList.append(roc_auc)
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title(used)
plt.plot([0, 1], [0, 1], '--', color=(0.6, 0.6, 0.6), label='Luck')
plt.legend()
plt.show()
for sample in aucList:
print(sample)
# plt.xlim([xmin,xmax])
# plt.ylim([xmin,xmax])
# plt.xticks([])
# plt.yticks([])
#plt.text(-60,20,'Disc loss: %f, NELBO: %f, KL0: %f, KL1: %f, KL2: %f, KL3: %f, KL4: %f' % (dl, NELBO, KL0, KL1, KL2, KL3, KL4))
#plt.text(-28,-7,'KLAVG: %f' %(np.mean([KL0,KL1,KL2,KL3,KL4])))
#plt.savefig('FiguresPCADV\Fig %i'%(j))
#plt.close()
np.savetxt("PCADV truklavg.csv", KLAVGBRUH, delimiter=",")
np.savetxt("PCADV estimatorloss.csv", ISTHISLOSS, delimiter=",")
np.savetxt("PCADV nelbo.csv", NELBRUH, delimiter=",")
np.savetxt("PCADV converge.csv", converge)
plt.plot(KLAVGBRUH)
plt.axvline(x=(converge[0] / 100))
plt.xlabel('Iterations (x100)')
plt.ylabel('Avg KL Div')
plt.savefig("PCADV truklavg")
plt.close()
plt.plot(ISTHISLOSS)
plt.axvline(x=converge[0])
plt.xlabel('Iterations')
plt.ylabel('Estimator Loss')
plt.savefig("PCADV estimatorloss")
plt.close()
plt.plot(NELBRUH)
plt.axvline(x=converge[0])
plt.xlabel('Iterations')
plt.ylabel('NELBO')
plt.savefig('PCADV nelbo')
plt.close()
#nice plot
def main():
#Load Data
dfGlucose = pd.read_csv(os.path.join('blood-glucose-data.csv'),
sep=',',
index_col='point_timestamp',
error_bad_lines=True)
dfActivity = pd.read_csv(os.path.join('distance-activity-data.csv'),
sep=',',
index_col='point_timestamp',
error_bad_lines=True)
dfHeart = pd.read_csv(os.path.join('heart-rate-data.csv'),
sep=',',
index_col='point_timestamp',
error_bad_lines=True)
#Convert point_value to numeric
dfGlucose['point_value(mg/dL)'] = dfGlucose['point_value(mg/dL)'].astype(
dtype=float)
dfActivity['point_value(kilometers)'] = dfActivity[
'point_value(kilometers)'].astype(dtype=float)
dfHeart['point_value'] = dfHeart['point_value'].astype(dtype=float)
#Convert index to time/data
dfGlucose.index = pd.to_datetime(dfGlucose.index)
dfActivity.index = pd.to_datetime(dfActivity.index)
dfHeart.index = pd.to_datetime(dfHeart.index)
#Aggregate data of Activity and Heart
#Activity
Activity = aggregate(dfGlucose.index, dfActivity,
'point_value(kilometers)', 'sum')
#Heart
Heart = aggregate(dfGlucose.index, dfHeart, 'point_value', 'mean')
#Plot and save the glucose values to explore Data
plt.figure()
plt.plot(dfGlucose.index,
dfGlucose['point_value(mg/dL)'].values,
label='Glucose')
pylab.xlabel('Timestamp')
pylab.ylabel('Glucose (mg/dL)')
plt.legend(loc=0)
plt.grid('on')
plt.savefig("Glucose_point_value(mg-dL).png")
plt.close()
#Plot and save the Activity values to explore Data
plt.figure()
plt.plot(dfGlucose.index, Activity, label='Aggregated Activity')
pylab.xlabel('Timestamp')
pylab.ylabel('Agg Activity')
plt.legend(loc=0)
plt.grid('on')
plt.savefig("Activity_aggregated.png")
plt.close()
#Plot and save the Heart values to explore Data
plt.figure()
plt.plot(dfGlucose.index, Heart, label='Aggregated Heart')
pylab.xlabel('Timestamp')
pylab.ylabel('Agg Heart')
plt.legend(loc=0)
plt.grid('on')
plt.savefig("Heart_aggregated.png")
plt.close()
#Separating instances of forecasting (discarting weeks with missing data)
'''For generate the whole set of slided data uncomment line 130 and comment line 131.
In order to save your time, the current configuration only generates data with the
Best valuated window sizes of the techniques. '''
#for window_size in range(48,150,4):
for window_size in range(52, 69, 4):
horizon = 12
#Transforming the data in instances of regression (univariate)
print('Sliding window_size: ' + str(window_size) + ' for glucose_only')
PipelineWindow(dfGlucose['point_value(mg/dL)'], [], window_size,
horizon, 'glucose_only')
print('Sliding window_size: ' + str(window_size) +
' for glucose_activity')
PipelineWindow(dfGlucose['point_value(mg/dL)'], [Activity],
window_size, horizon, 'glucose_activity')
print('Sliding window_size: ' + str(window_size) +
' for glucose_activity_heart')
PipelineWindow(dfGlucose['point_value(mg/dL)'], [Activity, Heart],
window_size, horizon, 'glucose_activity_heart')
def func(comp, param, trans_costs, critical_level, error_level):
dfa = df0[0:comp + param]
dfb = df00[0:comp + param]
#NOTE CRITICAL LEVEL HAS BEEN SET TO 5% FOR COINTEGRATION TEST
def find_cointegrated_pairs(dataframe, critical_level=0.05):
n = dataframe.shape[1] # the length of dateframe
pvalue_matrix = np.ones((n, n)) # initialize the matrix of p
keys = dataframe.columns # get the column names
pairs = [] # initilize the list for cointegration
for i in range(n):
for j in range(i + 1, n): # for j bigger than i
stock1 = dataframe[keys[i]] # obtain the price of "stock1"
stock2 = dataframe[keys[j]] # obtain the price of "stock2"
result = sm.tsa.stattools.coint(stock1,
stock2) # get conintegration
pvalue = result[1] # get the pvalue
pvalue_matrix[i, j] = pvalue
if pvalue < critical_level: # if p-value less than the critical level
pairs.append(
(keys[i], keys[j],
pvalue)) # record the contract with that p-value
return pvalue_matrix, pairs
#set up the split point for our "training data" on which to perform the co-integration test (the remaining data will be fed to our backtest function)
split = param
#run our dataframe (up to the split point) of ticker price data through our co-integration function and store results
pvalue_matrix, pairs = find_cointegrated_pairs(dfa[:-split],
critical_level)
def half_life(spread):
spread_lag = spread.shift(1)
spread_lag.iloc[0] = spread_lag.iloc[1]
spread_ret = spread - spread_lag
spread_ret.iloc[0] = spread_ret.iloc[1]
spread_lag2 = sm.add_constant(spread_lag)
model = sm.OLS(spread_ret, spread_lag2)
res = model.fit()
halflife = int(round(-np.log(2) / res.params[1], 0))
if halflife <= 0:
halflife = 1
return halflife
def backtest(dfa, dfb, param, s1, s2, error_level, trans_costs=False):
#############################################################
# INPUT:
# DataFrame of prices
# s1: the symbol of contract one
# s2: the symbol of contract two
# x: the price series of contract one
# y: the price series of contract two
# OUTPUT:
# df1['cum rets']: cumulative returns in pandas data frame
# sharpe: Sharpe ratio
# CAGR: Compound Annual Growth Rate
#Kalman filter params
delta = 1e-4
wt = delta / (1 - delta) * np.eye(2)
vt = 1e-3
theta = np.zeros(2)
C = np.zeros((2, 2))
R = None
# n = 4
# s1 = pairs[n][0]
# s2 = pairs[n][1]
x = list(dfa[s1])
y = list(dfa[s2])
ts_x = list(dfb[s1])
ts_y = list(dfb[s2])
#X = sm.add_constant(x)
# hrs = [0]*len(df)
Fs = [0] * len(dfa)
Rs = [0] * len(dfa)
ets = [0] * len(dfa)
Qts = [0] * len(dfa)
sqrt_Qts = [0] * len(dfa)
thetas = [0] * len(dfa)
Ats = [0] * len(dfa)
Cs = [0] * len(dfa)
for i in range(len(dfa) - param - 10, len(dfa)):
# res = sm.OLS(y[i-split+1:i+1],X[i-split+1:i+1]).fit()
# hr[i] = res.params[1]
F = np.asarray([x[i], 1.0]).reshape((1, 2))
Fs[i] = F
if R is not None:
R = C + wt
else:
R = np.zeros((2, 2))
Rs[i] = R
# Calculate the Kalman Filter update
# ----------------------------------
# Calculate prediction of new observation
# as well as forecast error of that prediction
yhat = F.dot(theta)
et = y[i] - yhat
ets[i] = et[0]
# Q_t is the variance of the prediction of
# observations and hence \sqrt{Q_t} is the
# standard deviation of the predictions
Qt = F.dot(R).dot(F.T) + vt
sqrt_Qt = np.sqrt(Qt)[0][0]
Qts[i] = Qt
sqrt_Qts[i] = sqrt_Qt
# The posterior value of the states \theta_t is
# distributed as a multivariate Gaussian with mean
# m_t and variance-covariance C_t
At = R.dot(F.T) / Qt
theta = theta + At.flatten() * et
C = R - At * F.dot(R)
thetas[i] = theta[0]
Ats[i] = At
Cs[i] = C
dfa['et'] = ets
# df['et'] = df['et'].rolling(5).mean()
dfa['sqrt_Qt'] = sqrt_Qts
dfa['theta'] = thetas
# run regression (including Kalman Filter) to find hedge ratio and then create spread series
df1 = pd.DataFrame({
'y': y,
'x': x,
'ts_x': ts_x,
'ts_y': ts_y,
'et': dfa['et'],
'sqrt_Qt': dfa['sqrt_Qt'],
'theta': dfa['theta']
})[-param:]
# df1[['et','sqrt_Qt']].plot()
# df1['spread0'] = df1['y'] - df1['theta']*df1['x']
# # calculate half life
# halflife = half_life(df1['spread0'])
# # calculate z-score with window = half life period
# meanSpread = df1.spread0.rolling(window=halflife).mean()
# stdSpread = df1.spread0.rolling(window=halflife).std()
# df1['zScore'] = (df1.spread0-meanSpread)/stdSpread
# #############################################################
# # Trading logic
# entryZscore = 0.8
# exitZscore = 0.2
#set up num units long
# df1['long entry'] = ((df1.zScore < - entryZscore) & (df1.zScore.shift(1) > - entryZscore))
# df1['long exit'] = ((df1.zScore > - exitZscore) & (df1.zScore.shift(1) < - exitZscore))
threshold = error_level * 0.001
df1['long entry'] = ((df1.et < -threshold) &
(df1.et.shift(1) > -threshold))
df1['long exit'] = ((df1.et > 0) & (df1.et.shift(1) < 0))
df1['num units long'] = np.nan
df1.loc[df1['long entry'], 'num units long'] = 1
df1.loc[df1['long exit'], 'num units long'] = 0
df1['num units long'][0] = 0
df1['num units long'] = df1['num units long'].fillna(method='pad')
#set up num units short
# df1['short entry'] = ((df1.zScore > entryZscore) & (df1.zScore.shift(1) < entryZscore))
# df1['short exit'] = ((df1.zScore < exitZscore) & (df1.zScore.shift(1) > exitZscore))
df1['short entry'] = ((df1.et > threshold) &
(df1.et.shift(1) < threshold))
df1['short exit'] = ((df1.et < 0) & (df1.et.shift(1) > 0))
df1.loc[df1['short entry'], 'num units short'] = -1
df1.loc[df1['short exit'], 'num units short'] = 0
df1['num units short'][0] = 0
df1['num units short'] = df1['num units short'].fillna(method='pad')
df1['numUnits'] = df1['num units long'] + df1['num units short']
df1['signals'] = df1['numUnits'].diff()
#df1['signals'].iloc[0] = df1['numUnits'].iloc[0]
df1['yfrets'] = df1['y'].pct_change().shift(-1)
df1['xfrets'] = df1['x'].pct_change().shift(-1)
if trans_costs == True:
df1['spread'] = (df1['y'] * (1 + df1['signals'] * df1['ts_y'])) - (
df1['theta'] * (df1['x'] * (1 - df1['signals'] * df1['ts_x'])))
#df1['spread'] = (df1['y']*(1+df1['signals']*0.0001))
else:
df1['spread'] = df1['y'] - df1['theta'] * df1['x']
#df1['spread'] = df1['y']
df1['spread pct ch'] = (df1['spread'] - df1['spread'].shift(1)) / (
(df1['x'] * abs(df1['theta'])) + df1['y'])
#df1['spread pct ch'] = (df1['spread'] - df1['spread'].shift(1)) / df1['spread'].shift(1)
df1['port rets'] = df1['spread pct ch'] * df1['numUnits'].shift(1)
df1['cum rets'] = df1['port rets'].cumsum()
df1['cum rets'] = df1['cum rets'] + 1
#df1 = df1.dropna()
##############################################################
try:
sharpe = ((df1['port rets'].mean() / df1['port rets'].std()) *
sqrt(252 * 2))
except ZeroDivisionError:
sharpe = 0.0
##############################################################
start_val = 1
end_val = df1['cum rets'].iloc[-1]
# print(len(df1[df1['long entry']==True])+len(df1[df1['short entry']==True]))
# print(end_val)
start_date = df1.iloc[0].name
end_date = df1.iloc[-1].name
days = (end_date - start_date).days
CAGR = round(((float(end_val) / float(start_val))**(252.0 / days)) - 1,
4)
df1[s1 + " " + s2] = df1['cum rets']
return df1[s1 + " " + s2], sharpe, CAGR
results = []
for pair in pairs:
rets, sharpe, CAGR = backtest(dfa, dfb, param, pair[0], pair[1],
trans_costs, error_level)
results.append(rets)
#print("The pair {} and {} produced a Sharpe Ratio of {} and a CAGR of {}".format(pair[0],pair[1],round(sharpe,2),round(CAGR,4)))
#rets.plot(figsize=(20,15),legend=True)
#concatenate together the individual equity curves into a single DataFrame
try:
results_df = pd.concat(results, axis=1).dropna()
#equally weight each equity curve by dividing each by the number of pairs held in the DataFrame
results_df /= len(results_df.columns)
#sum up the equally weighted equity curves to get our final equity curve
final_res = results_df.sum(axis=1)
#plot the chart of our final equity curve
plt.figure()
final_res.plot(figsize=(20, 15))
plt.title('Between {} and {}'.format(
str(results_df.index[0])[:10],
str(results_df.index[-1])[:10]))
plt.xlabel('Date')
plt.ylabel('Returns')
#calculate and print our some final stats for our combined equity curve
try:
sharpe = (final_res.pct_change().mean() /
final_res.pct_change().std()) * (sqrt(252 * 2))
except ZeroDivisionError:
sharpe = 0.0
start_val = 1
end_val = final_res.iloc[-1]
start_date = final_res.index[0]
end_date = final_res.index[-1]
days = (end_date - start_date).days
CAGR = round(((float(end_val) / float(start_val))**(252.0 / days)) - 1,
4)
print("Sharpe Ratio is {} and CAGR is {}".format(
round(sharpe, 2), round(CAGR, 4)))
except ValueError:
# return "No result"
print('No pair found')
import numpy as np
import sys
bins = np.linspace(-100, 100, 101)
std = np.sqrt(1000)
mu = 0.
def gauss(x, mu, sigma):
val = np.exp(-(x - mu)**2 / (2. * sigma**2))
val /= sigma * np.sqrt(2 * np.pi)
return val
vgauss = np.vectorize(gauss, excluded=['mu', 'sigma'])
for fl in sys.argv[1:]:
f = open(fl)
values = []
for line in f:
values.append(float(line))
plt.hist(values, bins=bins, histtype='step', normed=True)
plt.plot(bins, vgauss(bins, mu, std), color='black', lw=2)
plt.xlabel(r'Terminal Walk Position', fontsize=16)
plt.ylabel("Normalized Counts", fontsize=16)
plt.xticks(fontsize=16)
plt.yticks(fontsize=16)
plt.show()
import sys
import matplotlib.pylab as plt
f = open(sys.argv[1], 'r')
while True:
plot_name = f.readline()
if plot_name is None or len(plot_name) == 0:
break
plot_name = plot_name.rstrip()
fig, ax = plt.subplots()
while True:
label = f.readline().rstrip()
if label is None or len(label) == 0:
break
x = [float(n) for n in f.readline().rstrip().split(' ')]
y = [float(n) for n in f.readline().rstrip().split(' ')]
ax.loglog(x, y, label=label.rstrip())
ax.legend()
plt.title(plot_name)
plt.xlabel('Array size')
plt.ylabel('time (s)')
plot_name = plot_name.replace(':', '').replace(' ', '_')
plt.savefig(plot_name + ".png")
plt.close()
#%% Random Walk
import random
import matplotlib.pylab as plt
import numpy as np
def random_walk(n):
x = 0
c = [x]
for i in range(n):
prob = random.random()
if prob > 0.5:
x = x + 1
elif prob < 0.5:
x = x - 1
c.append(x)
return c
numberOfSteps = 100
walk = random_walk(numberOfSteps)
n = np.arange(numberOfSteps + 1)
plt.plot(n, walk, 'r-')
plt.plot(n, walk, 'bo')
plt.xlabel('Time (n)')
plt.ylabel('Position (x)')
plt.show()
#4.3.2 数值微分的例子
def function_1(x):
return 0.01 * x**2 + 0.1 * x
#画切线的函数
def tangent_line(f, x):
d = numerical_diff(f, x)
print(d)
y = f(x) - d * x #？？？
return lambda t: d * t + y
x = np.arange(0.0, 20.0, 0.1) #以0.1为间隔，从0到20
y = function_1(x)
plt.xlabel("x")
plt.ylabel("f(x)")
plt.plot(x, y)
plt.show()
print(numerical_diff(function_1, 5))
print(numerical_diff(function_1, 10))
tf = tangent_line(function_1, 5)
y2 = tf(x) #？？？
plt.plot(x, y)
plt.plot(x, y2) #切线
plt.show()
y4,
yerr=erry4,
fmt='r^',
ecolor='r',
markeredgecolor='red',
capthick=0.5,
label="TOTEM")
plt.fill_between(px3, py3, py4, color='c')
plt.fill_between(px5, py5, py6, color='c')
plt.plot(px1, py1, 'k-', linewidth=1.0, label=r"$pp$")
plt.plot(px2, py2, 'k--', linewidth=1.0, label=r"$\bar{p}p$")
plt.xlabel(r"$\sqrt{s}$ [GeV]", fontdict=font)
plt.ylabel(r"$\sigma_{tot}(s)$ [mb]", fontdict=font)
leg = plt.legend(loc=4,
ncol=1,
shadow=False,
fancybox=False,
frameon=False,
numpoints=1)
leg.get_frame().set_alpha(0.5)
###########################################
sub_axes = plt.axes([.2, .55, .35, .35])
plt.grid(False)
plt.semilogx()
plt.xlim(5000, 15000)
plt.ylim(90, 120)
import matplotlib.pylab as plt
data = genfromtxt("revised_AD_CNN_cross.csv", delimiter=",")
data[0, 0] = 0.93456
print(data[0, 0])
ax = sns.heatmap(data,
linewidth=0.01,
center=0.5,
cmap="PiYG",
cbar_kws={"ticks": [0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0]})
plt.xlabel("CNN model Index")
plt.ylabel("Target Index")
plt.tight_layout()
plt.savefig("revised_AD_cross_AUC.png", dpi=400)
plt.show()
import pandas as pd
data = pd.read_excel("revised_AD_CNN_cross.xlsx")
models = list(data.columns.values)
targets = list(data.index.values)
subsets = {
"GPCR":
rho2_systemic = rho42_005[200:]
system_error2= np.mean(rho2_systemic)
print "system error for dx=0.05 = ", system_error2
ratio = np.power((0.01/0.05),-2)
print "Compare expected ratio = ", ratio , "with computed ratio = " , system_error1/system_error2
plot1 =plt.plot(rho_42_005, color='green', label = r'$\Delta x=0.05$')
plot2 =plt.plot(rho_42_001, color ='blue', label = '$\Delta x=0.01$')
plt.legend(bbox_to_anchor=(1, 1), numpoints = 1 )
plt.xlabel('m', fontsize = 16)
plt.ylabel(r'$\Delta \rho$', fontsize = 20)
plt.annotate(r'$ \frac{\mathbb{E} \left[ \rho_{(\Delta x=0.01)} \right] }{\mathbb{E} \left[ \rho_{(\Delta x=0.05)} \right] }= 1.2028$', xy=(800, 0), xytext=(200, -0.1), fontsize=20)
plt.savefig("results/rho_particle42_ratio1p2.pdf")
plt.show()
plot1 =plt.plot(data1, color='green', label = r'$\Delta x=0.01$')
plot2 =plt.plot(data2, label = '$\Delta x=0.05$')
#plot1 =plt.plot(np.append(np.roll(value,1),value[9]) )
#plot1 =plt.plot( norm_Jv_fp )
#M=10
#norm_Jv_fp
#average_Jv_norm = scipy.zeros((M))
#for m in range(len(Jv_norm)):
fn = sArg[2:]
bFilename = True
# grid on
if "-G" in sArg:
bGrid = True
# filename is mandatory
if not bFilename:
sys.exit(
'Usage: python3 fpqSegmentAngleLengthCrossPlot.py -I<inputfilename>')
# alternative method
nodelist = fpq.getNodes(fn)
xnodelist = nodelist[0]
ynodelist = nodelist[1]
segangle = fpq.getSegAngles(xnodelist, ynodelist)
seglengths = fpq.getSegLengths(xnodelist, ynodelist)
nSegs = len(segangle)
# plot the segment length distribution
fig, ax = plt.subplots(figsize=(xSize, ySize))
plt.plot(segangle, seglengths, 'o')
plt.xlabel('Segment angle, degrees')
plt.ylabel('Segment length, pixels')
plt.xlim(0, 180)
plt.ylim(0, max(seglengths) * 1.05)
plt.grid(bGrid)
plt.title('Segment angles versus lengths, n=%i' % nSegs)
plt.savefig("fpqSegmentAngleLengthCrossPlot.png", dpi=600)
print('Plotted %5d segments, angles & lengths' % nSegs)
plt.text(0.245, yhaz2+yoff/5., '1/2475-year AEP', va='bottom',ha='right',fontsize=16)
plt.legend()
plt.grid(which='both')
# get x lims from haz curve 1
thaz = exp(interp(log(1e-4), log(sd1['poe_probs_annual'][::-1]), log(imls[::-1])))
# round to neares t 0.1
xmax = ceil(thaz / 0.1) * 0.1
plt.xlim([0, xmax])
plt.ylim([1e-4, .1])
plt.xlim([0, .25])
ax.tick_params(labelsize=14)
plt.ylabel('Annual Probabability of Exceedance', fontsize=16)
plt.xlabel(' '.join(('Mean', period, 'Hazard (g)')), fontsize=16)
# adjust x axes
#fig.subplots_adjust(hspace=0.2)
# save
if period == 'PGA' or period == 'PGV':
plt.savefig('_'.join(('nsha18_haz_curves', period +'.png')), format='png',bbox_inches='tight')
else:
plt.savefig('_'.join(('nsha18_haz_curves','SA('+period+').png')), format='png',bbox_inches='tight')
plt.show()
tick_params, xlim
#f = figure(figsize=(10, 3))
for i, folder in enumerate(folders):
if folder.startswith("B0"): #or folder.startswith("P0")) and \
if folder in ["B0010", "C0087", "C0093"]:
continue
print "Working on:", folder
case = path.join(basefolder, folder)
length, a = main(case, beta, smooth, stats, ratio)
# Plot
f = plot(length, a, linewidth=3, color=color[i])
hold("on")
ylabel("Area", fontsize=20)
#### FOR AREA VARIATION PLOT ####
tick_params(
axis='y', # changes apply to the y-axis
which='both', # both major and minor ticks are affected
left='off', # ticks along the bottom edge are off
right='off', # ticks along the top edge are off
labelleft='off') # labels along the bottom edge are off
tick_params(
axis='y', # changes apply to the y-axis
which='both', # both major and minor ticks are affected
top='off') # ticks
tight_layout()
x[0] = 0
exact_x[0] = 0
v[0] = 0
a[0] = 0
F[0] = 0
## Loop
for i in range(len(t) - 1):
b[i + 1] = b_C + u * t[i]
if (b[i] - x[i] - b_C) > (m * g * fr_s) / k:
a[i + 1] = (k / m) * (b[i] - x[i] - b_C) - g * fr_d
v[i + 1] = v[i] + dt * a[i + 1]
F[i + 1] = k * (b[i] - x[i] - b_C)
else:
a[i + 1] = 0
v[i + 1] = 0
F[i + 1] = 0
x[i + 1] = x[i] + dt * v[i + 1]
exact_x[i + 1] = u * t[i] - (u / w) * np.sin(w * t[i])
plt.plot(t, F)
plt.xlabel("t")
plt.ylabel("Y")
plt.title("m = 1.0")
plt.legend(["Force", "Velocity", "Acceleration"])
#plt.savefig("oppg_t_1.png")
plt.show()
eva_mean = eigvals_col.mean(axis=0)
data = np.load(join(figdir, "H_avg_1000cls.npz"))
eva_Havg, evc_Havg, Havg = data['eigvals_avg'], data['eigvects_avg'], data['H_avg']
#%% BigGAN
plt.figure(figsize=[5, 4])
eva00 = np.load(join(datadir, "eig_gen_z.npz"))["eva"] #, H=H00, eva=eva00, evc=evc00)
plt.plot(np.log10(eva00 / eva00.max())[::-1], label="gen_z")
for blocki in range(Ln):
eva00 = np.load(join(datadir, "eig_genBlock%02d.npz" % blocki))["eva"]
plt.plot(np.log10(eva00 / eva00.max())[::-1], color=cm.jet((blocki+1) / (Ln+1)),
label=("GenBlock%02d" % blocki) if blocki!=8 else "SelfAttention")
plt.plot(np.log10(eva_mean / eva_mean.max()), color=cm.jet((Ln+1) / (Ln+1)),
label="GAN")
plt.xlim([0, len(eva00)])
plt.xlabel("eigenvalue rank")
plt.ylabel("log(eig / eig max)")
plt.title("BigGAN Hessian Spectra of\n Intermediate Layers Compared")
plt.subplots_adjust(top=0.9)
plt.legend()
plt.savefig(join(figdir, "spectrum_med_Layers_norm_cmp.png"))
plt.savefig(join(figdir, "spectrum_med_Layers_norm_cmp.pdf"))
plt.show()
#%% Collect the H and eva evc into list.
npzlabel = ["gen_z"]+["genBlock%02d"%blocki for blocki in range(Ln)]
layernames = ["gen_z"] + [("GenBlock%02d" % blocki) if blocki!=8 else "SelfAttention" for blocki in range(Ln)] \
+ ["Full"]
H_col = []
evc_col = []
eva_col = []
for label in npzlabel:
data = np.load(join(datadir, "eig_%s.npz" % label))
print elapsedtime, ' seconds'
# ~ ~ ~ ~ Plot Results ~ ~ ~ ~ ~ ~ #
lambd = x[3] * pixels**2 + x[4] * pixels + x[5]
plt.figure()
lsf = sumgaus(x, spec[:, 0], nsat)
#lsf = gaussian(spec[:,0],sig)
tau = x[0]
norm = x[1]
sig = x[2]
dxa = x[3]
dxb = x[4]
dxc = x[5]
amps = x[6:len(x)]
model = fftconvolve(norm * (spec[:, 1]**tau), lsf, 'same')
modelinterp = interp1d(spec[:, 0], model)
plt.plot(lambd, realspec0, 'k', lambd, modelinterp(lambd), 'g')
plt.figure()
plt.plot(lambd, realspec0 - modelinterp(lambd), 'g')
plt.figure()
plt.plot(chi_ev)
plt.xlabel('N_iteration')
plt.ylabel('Function Value')
#best one
#array([ 3.48537391e-01, 9.88941209e-01, 1.38819216e-02,
# -3.95332754e-07, 1.10756971e-02, 8.16905624e+02])
import numpy as np
import matplotlib.pylab as plt
import matplotlib as mp
from matplotlib.backends.backend_pdf import PdfPages
from scipy.stats import beta
# mp.rc('font', family = 'serif', serif = 'cmr10')
mp.rcParams['mathtext.fontset'] = 'cm'
mp.rcParams.update({'font.size': 16})
a = 2
b = 2
x = np.linspace(0,1,1000)
u = np.ones(len(x))
u[0] = 0
u[-1] = 0
with PdfPages('distance_distributions.pdf') as pdf:
plt.plot(0.2*x,5*beta.pdf(x, a, b), label="paired distribution")
plt.plot(x,u, label="unpaired distribution")
plt.xlabel("distance, $d$")
plt.ylabel("probability density")
plt.legend(loc="best")
pdf.savefig(bbox_inches="tight")
plt.yticks(np.linspace(-2, 0, 5),fontsize=20)
plt.rc('xtick',labelsize=15)
plt.rc('ytick',labelsize=15)
ax = plt.gca()
ax.spines['top'].set_color('none')
ax.spines['right'].set_color('none')
ax.xaxis.set_ticks_position('bottom')
ax.yaxis.set_ticks_position('left')
ax.spines['left'].set_position(('data',0))
ax.spines['bottom'].set_position(('data',0))
plt.legend(loc=4,fontsize=13)
plt.title('rho = 60um, w0 = 10um',fontsize=20)
plt.xlabel('w/rho',fontsize=20)
plt.ylabel('Qz',fontsize=20)
plt.grid()
plt.show()
MTP.table_parameter('x-axis w0 to 7rho', 'related to w', '0','0, rho, 2rho, 3rho', '30')
t_finish = time.time()
print(f"script executed in {t_finish-t_start:.2f} seconds")
'''
if pops[-1] < 0:
RC0[-1, -1] = 0
print("Predator reached extinction after {} iterations".format(i))
break
# ensure time is not longer than the RC0 data
t_axis = range(len(RC0))
# plot it
f1 = p.figure()
p.plot(t_axis, RC0[:, 0], 'g-', label='Resource density') # Plot
p.plot(t_axis, RC0[:, 1], 'b-', label='Consumer density')
p.grid()
p.legend(loc='best')
p.xlabel('Time')
p.ylabel('Population density')
p.title('Consumer-Resource population dynamics')
# save the figure as a pdf`
f1.savefig('../Results/Plots/LV3_model.pdf')
# plot the second circular figure. this just has the values from above as axis
f2 = p.figure()
p.plot(RC0[:, 0], RC0[:, 1], 'r-') # Plot
p.grid()
p.legend(loc='best')
p.xlabel('Resource density')
p.ylabel('Consumer density')
p.title('Consumer-Resource population dynamics')
# save as pdf
from pylab import boxplot
warnings.filterwarnings("ignore")
pvalue = []
#Read the csv file and get only the time series
series = pd.read_csv('C:\Git\sdl\code\dados\lag15Min-BVMF3.csv')
series.drop('Unnamed: 0', axis=1, inplace=True)
series.drop('INSTRUMENT', axis=1, inplace=True)
series.drop('TRADE_TIME', axis=1, inplace=True)
print(series.head())
plt.plot(series)
plt.title('PETR4 May fluctuation - lag 15min')
plt.xlabel('Time (min)')
plt.ylabel('Price (Brazilian Reais)')
plt.show()
#Plot correlation and autocorrelation plots to analyze how the numbers relate
print('ACF and PACF for the original series')
pyplot.figure()
plot_acf(series,
ax=pyplot.gca(),
lags=20,
title='Autocorrelation plot for original series')
pyplot.show()
pyplot.figure()
plot_pacf(series,
ax=pyplot.gca(),
lags=20,
def plot_traces(traces, fs, n, dataset, filename, rand=True, pre_traces=None):
# Data len
N = traces.shape[1]
# Time axis for signal plot
t_ax = np.arange(N) / fs
# Frequency axis for FFT plot
xf = np.linspace(-fs / 2.0, fs / 2.0 - 1 / fs, N)
# Traces to plot
trtp = []
if rand:
# Init rng
rng = default_rng()
# Traces to plot numbers
trtp_ids = rng.choice(len(traces), size=n, replace=False)
trtp_ids.sort()
# Retrieve selected traces
for idx, trace in enumerate(traces):
if idx in trtp_ids:
trtp.append(trace)
else:
trtp_ids = pre_traces
# Retrieve selected traces
for idx, trace in enumerate(traces):
if idx in trtp_ids:
trtp.append(trace)
# Plot traces in trtp with their spectrum
for idx, trace in enumerate(trtp):
yf = sfft.fftshift(sfft.fft(trace))
gs = gridspec.GridSpec(2, 2)
pl.figure()
pl.subplot(gs[0, :])
pl.plot(t_ax, trace)
pl.title(
f'Traza {dataset} y espectro #{trtp_ids[idx]} archivo {filename}')
pl.xlabel('Tiempo [s]')
pl.ylabel('Amplitud [-]')
pl.grid(True)
pl.subplot(gs[1, 0])
pl.plot(xf, np.abs(yf) / np.max(np.abs(yf)))
pl.xlabel('Frecuencia [Hz]')
pl.ylabel('Amplitud [-]')
pl.grid(True)
pl.subplot(gs[1, 1])
pl.plot(xf, np.abs(yf) / np.max(np.abs(yf)))
pl.xlim(-25, 25)
pl.xlabel('Frecuencia [Hz]')
pl.ylabel('Amplitud [-]')
pl.grid(True)
pl.tight_layout()
pl.savefig(f'Imgs/{dataset}/{filename}/{trtp_ids[idx]}.png')
def run_opt(self, popsize, numgen, processors,
plot=False, log=False, **kwargs):
"""
Runs the optimizer.
:param popsize:
:param numgen:
:param processors:
:param plot:
:param log:
:param kwargs:
:return:
"""
self._params['popsize'] = popsize
self._params['numgen'] = numgen
self._params['processors'] = processors
self._params['plot'] = plot
self._params['log'] = log
# allows us to pass in additional arguments e.g. neighbours
self._params.update(**kwargs)
self.halloffame = tools.HallOfFame(1)
self.stats = tools.Statistics(lambda thing: thing.fitness.values)
self.stats.register("avg", numpy.mean)
self.stats.register("std", numpy.std)
self.stats.register("min", numpy.min)
self.stats.register("max", numpy.max)
self.logbook = tools.Logbook()
self.logbook.header = ["gen", "evals"] + self.stats.fields
self._params['model_count'] = 0
start_time = datetime.datetime.now()
self.initialize_pop()
for g in range(self._params['numgen']):
self.update_pop()
self.halloffame.update(self.population)
self.logbook.record(gen=g, evals=self._params['evals'],
**self.stats.compile(self.population))
print(self.logbook.stream)
end_time = datetime.datetime.now()
time_taken = end_time - start_time
self._params['time_taken'] = time_taken
print("Evaluated {0} models in total".format(
self._params['model_count']))
print("Best fitness is {0}".format(self.halloffame[0].fitness))
print("Best parameters are {0}".format(self.parse_individual(
self.halloffame[0])))
for i, entry in enumerate(self.halloffame[0]):
if entry > 0.95:
print(
"Warning! Parameter {0} is at or near maximum allowed "
"value\n".format(i + 1))
elif entry < -0.95:
print(
"Warning! Parameter {0} is at or near minimum allowed "
"value\n".format(i + 1))
if self._params['log']:
self.log_results()
if self._params['plot']:
print('----Minimisation plot:')
plt.figure(figsize=(5, 5))
plt.plot(range(len(self.logbook.select('min'))),
self.logbook.select('min'))
plt.xlabel('Iteration', fontsize=20)
plt.ylabel('Score', fontsize=20)
]
#plt.plot(x,y)
#plt.show()
###################3
u = np.linspace(-1, 1.5, 50)
v = np.linspace(-1, 1.5, 50)
z = np.zeros((len(u), len(v)))
def mapFeatureForPlotting(X1, X2):
degree = 6
out = np.ones(1)
for i in range(1, degree + 1):
for j in range(i + 1):
out = np.hstack(
(out, np.multiply(np.power(X1, i - j), np.power(X2, j))))
return out
for i in range(len(u)):
for j in range(len(v)):
z[i, j] = np.dot(mapFeatureForPlotting(u[i], v[j]), theta.x)
plt.contour(u, v, z, 0)
plt.xlabel('Microchip Test1')
plt.ylabel('Microchip Test2')
#plt.legend((passed, failed), ('Passed', 'Failed'))
plt.show()
model.add(Flatten())
model.add(Dense(1000, activation='relu'))
model.add(Dense(num_classes, activation='softmax'))
model.compile(loss=keras.losses.categorical_crossentropy,
optimizer=keras.optimizers.Adam(),
metrics=['accuracy'])
class AccuracyHistory(keras.callbacks.Callback):
def on_train_begin(self, logs={}):
self.acc = []
def on_epoch_end(self, batch, logs={}):
self.acc.append(logs.get('acc'))
history = AccuracyHistory()
model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_data=(x_test, y_test),
callbacks=[history])
score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])
plt.plot(range(1, 11), history.acc)
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
plt.show()