def scatter_plot(P, L, pcIdx1, pcIdx2, letterList, rev):
fig = plt.figure()
# following the convention in lecture note ScatterPlot.html
colors = ["r", "lime", "b", "y", "c", "m", "k", "tan", "pink", "darkred"]
for i, letter in enumerate(letterList):
plt.scatter(P[L == letter, pcIdx2],
P[L == letter, pcIdx1],
s=0.1,
c=colors[i],
label=letter)
plt.axes().set_aspect('equal')
#plt.axes().set_aspect('equal', 'datalim')
plt.xlabel("Principle Component {}".format(pcIdx2))
plt.ylabel("Principle Component {}".format(pcIdx1))
plt.axhline(0, color='grey')
plt.axvline(0, color='grey')
plt.ylim([-5000, 5000])
plt.xlim([-5000, 5000])
plt.legend()
plt.gca().invert_yaxis()
fig.set_size_inches(8, 8)
fName = os.path.join(
pDir, 'scatter_PC{}_PC{}_{}_{}.png'.format(pcIdx1, pcIdx2,
"".join(letterList), rev))
savefig(fName, bbox_inches='tight')
plt.show()
def showModeless(self):
fig = figure(1)
ax = axes([0.025, 0.5-len(self.checkStates)*.05*.5, 0.14, len(self.checkStates)*.05])
self.yBtns = CheckButtons(ax, self.names, [self.checkStates[name] for name in self.names])
self.yBtns.on_clicked(self.onItemselected)
axes([0.2, 0.1, 0.75, 0.85])
self.plot()
manager = get_current_fig_manager()
self.loadSetting(manager.window)
self.registerCallbacks(fig, manager.window)
def Plot_Circle(c_list):
plt.figure()
plt.axes().set_aspect('equal')
plt.xlim([-1, 1])
plt.ylim([-1, 1])
theta = np.linspace(0, 2 * np.pi, 90)
for c in c_list:
x = c.x
y = c.y
r = c.radius
plt.plot(x + r * np.cos(theta), y + r * np.sin(theta), 'r')
plt.show()
def showModeless(self):
fig = figure(1)
ax = axes([
0.025, 0.5 - len(self.checkStates) * .05 * .5, 0.14,
len(self.checkStates) * .05
])
self.yBtns = CheckButtons(
ax, self.names, [self.checkStates[name] for name in self.names])
self.yBtns.on_clicked(self.onItemselected)
axes([0.2, 0.1, 0.75, 0.85])
self.plot()
manager = get_current_fig_manager()
self.loadSetting(manager.window)
self.registerCallbacks(fig, manager.window)
def feature_PCA():
my_arrays = load_data(0, get_avg=True)
for i in range(1, 89):
my_arrays = np.concatenate((my_arrays, load_data(i, get_avg=True)),
axis=0)
my_tensor = torch.from_numpy(my_arrays)
my_labels = np.loadtxt('hist/rotation.txt')
color_choice = ['b', 'g', 'r', 'yellow']
pca = PCA(n_components=3)
x_numpy = my_tensor.data.cpu().numpy()
x_pca = pca.fit_transform(x_numpy)
plt.figure()
ax = plt.axes(projection='3d')
for i in range(x_pca.shape[0]):
color_index = int(my_labels[i])
# plt.plot(x_pca[i, 0], x_pca[i, 1], 'o', color=color_choice[color_index])
ax.scatter(x_pca[i, 0],
x_pca[i, 1],
x_pca[i, 2],
color=color_choice[color_index])
# plt.text(X_norm[i, 0], X_norm[i, 1], str(y[i]), color=plt.cm.Set1(y[i]),
# fontdict={'weight': 'bold', 'size': 9})
ax.set_xlabel('X Label')
ax.set_ylabel('Y Label')
ax.set_zlabel('Z Label')
plt.show()
def __save(self,n,plot,sfile):
p.figure(figsize=sfile)
p.xlabel(plot.xlabel)
p.ylabel(plot.ylabel)
p.xscale(plot.xscale)
p.yscale(plot.yscale)
p.grid()
for curve in plot.curves:
if curve[1] == None: p.plot(curve[0],curve[2], label=curve[3])
else: p.plot(curve[0], curve[1], curve[2], label=curve[3])
p.rc('legend', fontsize='small')
p.legend(shadow=0, loc='best')
p.axes().set_aspect(plot.aspect)
if not plot.dir: plot.dir = './plots/'
if not plot.name: plot.name = self.__global_name+'_%0*i'%(2,n)
if not os.path.isdir(plot.dir): os.mkdir(plot.dir)
if plot.pgf: p.savefig(plot.dir+plot.name+'.pgf')
else: p.savefig(plot.dir+plot.name+'.pdf', bbox_inches='tight')
p.close()
def __init__(self):
QWidget.__init__(self)
self.fig = plt.figure()
self.canvas = FigureCanvas(self.fig)
self.ax = plt.axes(projection='3d')
self.affichageCourbe('Maillage\Rectangular_HULL_Normals_Outward.stl')
self.canvas.draw()
layout = QVBoxLayout()
layout.addWidget(self.canvas)
self.setLayout(layout)
def dimension_reduction(method, dimension):
my_arrays = load_data(0, get_avg=True)
for i in range(1, 89):
my_arrays = np.concatenate((my_arrays, load_data(i, get_avg=True)),
axis=0)
# print(my_arrays.shape)
my_tensor = torch.from_numpy(my_arrays)
my_labels = np.loadtxt('hist/rotation.txt')
print(my_labels.shape)
# time.sleep(30)
color_choice = ['b', 'g', 'r', 'yellow']
# print(my_tensor)
if method == 'tsne':
module = manifold.TSNE(n_components=dimension,
init='random',
random_state=500,
early_exaggeration=5,
method='exact')
elif method == 'PCA':
module = PCA(n_components=3)
x_numpy = my_tensor.data.cpu().numpy()
x_reduced = module.fit_transform(x_numpy)
print(x_reduced.shape)
print(x_reduced)
plt.figure()
if dimension == 2:
for i in range(x_reduced.shape[0]):
color_index = int(my_labels[i])
plt.plot(x_reduced[i, 0],
x_reduced[i, 1],
'o',
color=color_choice[color_index])
plt.xticks()
plt.yticks()
elif dimension == 3:
ax = plt.axes(projection='3d')
for i in range(x_reduced.shape[0]):
color_index = int(my_labels[i])
ax.scatter(x_reduced[i, 0],
x_reduced[i, 1],
x_reduced[i, 2],
color=color_choice[color_index])
ax.set_xlabel('X Label')
ax.set_ylabel('Y Label')
ax.set_zlabel('Z Label')
plt.savefig('plots/{}-24bin-{}d.pdf'.format(method, dimension))
plt.show()
def feature_visualization(method, dimension):
color_choice = ['b', 'g', 'r', 'yellow']
x_numpy = np.loadtxt('experiments/features.txt')
my_labels = np.loadtxt('experiments/features_labels.txt')
print(x_numpy.shape)
time.sleep(30)
if method == 'tsne':
module = manifold.TSNE(n_components=dimension,
init='pca',
random_state=500,
early_exaggeration=20,
method='exact')
elif method == 'PCA':
module = PCA(n_components=3)
x_reduced = module.fit_transform(x_numpy)
print(x_reduced.shape)
print(x_reduced)
plt.figure()
if dimension == 2:
for i in range(x_reduced.shape[0]):
color_index = int(my_labels[i])
if my_labels[i] == 0:
status = 'Died'
else:
status = 'Survived'
plt.plot(x_reduced[i, 0],
x_reduced[i, 1],
'o',
color=color_choice[color_index],
label=status)
plt.xticks()
plt.yticks()
elif dimension == 3:
ax = plt.axes(projection='3d')
for i in range(x_reduced.shape[0]):
color_index = int(my_labels[i])
ax.scatter(x_reduced[i, 0],
x_reduced[i, 1],
x_reduced[i, 2],
color=color_choice[color_index])
ax.set_xlabel('X Label')
ax.set_ylabel('Y Label')
ax.set_zlabel('Z Label')
plt.legend(loc="lower right")
plt.savefig('experiments/{}_{}.pdf'.format(method, dimension))
plt.show()
def plot_cov(qstack):
fig = plt.figure()
ax = plt.axes(None, label=str(qstack.bin_size))
flux_corr = np.corrcoef(qstack.fs_boot, rowvar=False)
flux_covar = qstack.flux_covar
im = ax.pcolormesh(qstack.wave_grid,
qstack.wave_grid,
flux_corr,
vmin=-1,
vmax=1)
plt.ylim(qstack.wave_grid[-1], qstack.wave_grid[0])
cbar = fig.colorbar(im, ax=ax)
cbar.set_label('Correlation')
#plt.title("Continuum Normalized Flux Covariance Matrix Across Wavelength Bins")
plt.xlabel("Wavelength Bin (Angstroms)")
plt.ylabel("Wavelength Bin (Angstroms)")
def plot_std(qstack):
plt.figure()
flux_covar = qstack.flux_covar
std = np.sqrt(np.diagonal(flux_covar))
ax = plt.axes(None, label=str(bin_size))
plt.plot(qstack.wave_stack, qstack.flux_stack, label='Stacked Flux')
ax.fill_between(qstack.wave_stack,
qstack.flux_stack - std,
qstack.flux_stack + std,
alpha=0.25,
label="1-$\\sigma$ Uncertainty Range")
#plt.title("Stacked Continuum Normalized Flux Near Ly-$\\alpha$ Transition")
plt.xlabel("Wavelength (Angstroms)")
plt.ylabel("Stacked Continuum Normalized Flux")
plt.axvline(x=1215.67, color='red', linestyle='--')
plt.legend()
def glm_scatter(x,
y,
pdf_path=None,
title=None,
x_label='x',
y_label='y',
formula='y ~ x',
show_sum=False):
"""Creates a scatter plot with a GLM model running through the data
:param x:
:param y:
:param pdf_name:
:param title:
:param x_label:
:param y_label:
:param formula:
:param show_sum:
:return:
"""
# Make the plotting grid
a = plt.axes(title=title)
a.set_xlabel(x_label)
a.set_ylabel(y_label)
# Prepare data and model
scatter_df = pd.DataFrame({'x': x, 'y': y}).sort_values('x')
mod = smf.ols(formula=formula, data=scatter_df).fit()
scatter_df['pred'] = mod.predict(scatter_df['x'])
scatter_df.plot('x', 'y', kind='scatter', ax=a)
scatter_df.plot('x', 'pred', ax=a, color='r')
# Show plot
plt.show()
if pdf_path:
pdf_path = add_filetype(pdf_path, '.pdf')
with PdfPages(pdf_path) as pdf:
plt.savefig(pdf, format='pdf')
plt.close('all')
if show_sum:
print(mod.summary())
breaks()
def plot_boot(qstack):
plt.figure()
flux_covar = qstack.flux_covar
std = np.sqrt(np.diagonal(flux_covar))
ax = plt.axes(None, label=str(bin_size))
num = 100
ws_boot = qstack.ws_boot[:100]
fs_boot = qstack.fs_boot[:100]
plt.plot(ws_boot.T, fs_boot.T, alpha=0.1, color='orange')
plt.plot(ws_boot[0],
fs_boot[0],
alpha=0.1,
color='orange',
label='Bootstrap Samples')
plt.plot(qstack.wave_stack, qstack.flux_stack, label='Stacked Flux')
#plt.title("Stacked Continuum Normalized Flux Near Ly-$\\alpha$ Transition")
plt.xlabel("Wavelength (Angstroms)")
plt.ylabel("Stacked Continuum Normalized Flux")
plt.axvline(x=1215.67, color='red', linestyle='--')
plt.legend()
fluo_paths = []
for i in xrange(len(paths)):
emissions = [v_inf_mean[t] for t in paths[i]]
f_cp = np.convolve(kernel[::-1], emissions, mode='full')
f_cp = f_cp[w:-w + 1]
fluo_paths.append(f_cp)
#plt.plot(np.array(fluo_paths[-1]))
#plt.plot(fluo_states[0])
#plt.show()
#--------------------------------------Generate Animation--------------------------------------------------------------#
# First set up the figure, the axis, and the plot element we want to animate
fig = plt.figure()
ax = plt.axes(xlim=(0, len(fluo_states[0])),
ylim=(0, np.max(fluo_states[0])))
ax.plot(range(len(fluo_states[0])), fluo_states[0])
line, = ax.plot([], [], lw=4)
# initialization function: plot the background of each frame
def init():
line.set_data([], [])
return line,
# animation function. This is called sequentially
def animate(i):
y = np.array(fluo_paths[i])
x = np.arange(0, len(y))
line.set_data(x, y)
return line,
s = Series(np.random.randn(10).cumsum(),index=np.arange(0,100,10))
s.plot()
plt.save('Figure_14.png')
df
df = DataFrame(np.random.randn(10,4).cumsum(0),
columns=['A','B','C','D'],
index=np.arange(0,100,10))
df.plot
df = DataFrame(np.random.randn(10,4).cumsum(0),
columns=['A','B','C','D'],
index=np.arange(0,100,10))
df.plot()
plt.axes
plt.axes()
import matplotlib.pyplot as plt
fig, axes = plt.subplots(2,1)
data = Series(np.random.randn(16),index=list('abcdefghijklmnop'))
data.plot(kind='bar', ax=axes[0],color='k', alpha=0.7)
data = Series(np.random.rand(16),index=list('abcdefghijklmnop'))
data.plot(kind='bar', ax=axes[0],color='k', alpha=0.7)
fig, axes = plt.subplots(2,1)
data = Series(np.random.rand(16),index=list('abcdefghijklmnop'))
data.plot(kind='bar', ax=axes[0],color='k', alpha=0.7)
data.plot(kind='barh', ax=axes[1],color='k', alpha=0.7)
fig.savefig('Figure_16.png')
df = DataFrame(np.random.rand(6,4),)
df = DataFrame(np.random.rand(6,4),
index=['r1','r2','r4','r5','r6'],
columns=pd.Index(['A','B','C','D'],name='Genus'))
chetty.e_rank_b.plot.hist()
# In[22]:
chetty.shape
# In[23]:
low_100 = chetty.sort_values(by="e_rank_b").head(100)
high_100 = chetty.sort_values(by="e_rank_b", ascending=False).head(100)
# In[25]:
imagery = img_tiles.GoogleTiles()
ax = mpl.axes(projection=imagery.crs)
#x0, x1, y0, y1
maps_limits = (-71.38, -70.77, 42.03, 42.47)
ax.set_extent(maps_limits)
ax.add_image(imagery, 10)
mpl.plot(low_100.LON,
low_100.LAT,
transform=ccrs.Geodetic(),
marker='.',
markersize=10,
color="red",
linewidth=0,
alpha=0.5)
plot(x, Psi0.conj().imag, 'y--', label=r"${\Psi_{0}}_{imag}$")
legend()
plt.show()
# In[3]:
from matplotlib import pyplot as plt
import numpy as np
from matplotlib.animation import FuncAnimation
# initializing a figure in
# which the graph will be plotted
fig = plt.figure()
# marking the x-axis and y-axis
axis = plt.axes(xlim=(-50, 50), ylim=(-0.5, 1))
plt.plot(x, v(x) * 0.1)
# initializing a line variable
line = axis.plot(x, p[0].real**2 + p[0].imag**2, lw=2)[0]
# data which the line will
# contain (x, y)
def init():
line.set_data([], [])
return line
def animate(i):
# plots a sine graph
line.set_ydata(p[i].real**2 + p[i].imag**2)
# In[33]:
plotter.plot({'Basic': history}, metric="mae")
plt.ylabel('MAE [Deaths]')
# In[34]:
plotter.plot({'Basic': history}, metric="mse")
plt.ylabel('MSE [Deaths^2]')
# In[35]:
test_predictions = model.predict(normed_test_data).flatten()
a = plt.axes(aspect='equal')
plt.scatter(test_labels, test_predictions)
plt.xlabel('True Values [Deaths]')
plt.ylabel('Predictions [Deaths]')
# In[36]:
error = test_predictions - test_labels
plt.hist(error, bins=25)
plt.xlabel("Prediction Error [Deaths]")
_ = plt.ylabel("Count")
# In[37]:
import plotly
import plotly.figure_factory as ff
def feature_visualization_remake(vectors, labels, method, dimension, figname):
class_num = int(np.max(labels) + 1)
print('there are {} classes'.format(class_num))
# time.sleep(30)
color_choice = ['g', 'r', 'b', 'yellow']
x_numpy = vectors
my_labels = labels
print('my_labels {}'.format(my_labels))
my_labels = np.reshape(my_labels, (my_labels.shape[0], 1))
mydata = np.concatenate((my_labels, x_numpy), axis=1)
np.random.shuffle(mydata)
x_numpy = mydata[:, 1:]
my_labels = mydata[:, 0]
print('Size of vectors {}'.format(x_numpy.shape))
print('Size of labels {}'.format(my_labels.shape))
if method == 'tsne':
module = manifold.TSNE(n_components=dimension,
init='random',
random_state=10,
early_exaggeration=20,
method='barnes_hut')
elif method == 'pca':
module = PCA(n_components=100)
x_reduced = module.fit_transform(x_numpy)
plt.figure()
plt.xticks([])
plt.yticks([])
plt.legend(loc="upper right")
plt.tight_layout()
if dimension == 2:
for class_index in range(0, class_num):
this_index = np.where(my_labels == class_index)
class_vectors = x_reduced[this_index]
plt.plot(class_vectors[:, 0],
class_vectors[:, 1],
'o',
color=color_choice[class_index],
alpha=0.5,
label='Class {}'.format(class_index))
if class_index == 1:
plt.legend(loc="upper right")
plt.savefig('experiments/{}_{}_{}_2.png'.format(
method, dimension, figname))
elif dimension == 3:
ax = plt.axes(projection='3d')
for class_index in range(0, class_num):
this_index = np.where(my_labels == class_index)
class_vectors = x_reduced[this_index]
ax.scatter(class_vectors[:, 0],
class_vectors[:, 1],
class_vectors[:, 2],
color=color_choice[class_index],
label='Class {}'.format(class_index))
ax.set_xlabel('X Label')
ax.set_ylabel('Y Label')
ax.set_zlabel('Z Label')
# plt.title('t-SNE Visualization of Patch-34 Feature Vectors')
# plt.legend(loc="upper right")
# plt.savefig('experiments/{}_{}_{}.pdf'.format(method, dimension))
plt.savefig('experiments/{}_{}_{}_3.png'.format(method, dimension,
figname))
def feature_visualization2(method, dimension, point_num=1700):
color_choice = ['b', 'g', 'r', 'yellow']
x_numpy = np.loadtxt('data/gt_vectors.txt')
my_labels = np.loadtxt('data/gt_labels.txt')
print('my_labels {}'.format(my_labels))
my_labels = np.reshape(my_labels, (my_labels.shape[0], 1))
mydata = np.concatenate((my_labels, x_numpy), axis=1)
np.random.shuffle(mydata)
x_numpy = mydata[:, 1:]
my_labels = mydata[:, 0]
x_numpy = x_numpy[:point_num, :]
my_labels = my_labels[:point_num]
print('Size of vectors {}'.format(x_numpy.shape))
print('Size of labels {}'.format(my_labels.shape))
# print('labels {}'.format(my_labels))
died_index = np.where(my_labels == 0)
survived_index = np.where(my_labels == 1)
if method == 'tsne':
module = manifold.TSNE(n_components=dimension,
init='pca',
random_state=10,
early_exaggeration=20,
method='barnes_hut')
elif method == 'pca':
module = PCA(n_components=100)
x_reduced = module.fit_transform(x_numpy)
died_vector = x_reduced[died_index]
survived_vector = x_reduced[survived_index]
plt.figure()
if dimension == 2:
plt.plot(died_vector[:, 0],
died_vector[:, 1],
'o',
color='tomato',
alpha=0.5,
label='Deceased')
plt.plot(survived_vector[:, 0],
survived_vector[:, 1],
'o',
color='limegreen',
alpha=0.5,
label='Survived')
# plt.plot(x_reduced[i, 0], x_reduced[i, 1], 'o', color=color_choice[color_index])
plt.xticks([])
plt.yticks([])
elif dimension == 3:
ax = plt.axes(projection='3d')
ax.scatter(died_vector[:, 0],
died_vector[:, 1],
died_vector[:, 2],
color='r',
label='Died')
ax.scatter(survived_vector[:, 0],
survived_vector[:, 1],
survived_vector[:, 2],
color='g',
label='Survived')
ax.set_xlabel('X Label')
ax.set_ylabel('Y Label')
ax.set_zlabel('Z Label')
# plt.title('t-SNE Visualization of Patch-34 Feature Vectors')
plt.legend(loc="upper right")
plt.tight_layout()
plt.savefig('experiments/{}_{}.pdf'.format(method, dimension))
plt.savefig('experiments/{}_{}.png'.format(method, dimension))
plt.show()
n_redundant=0,
random_state=42)
train_X, test_X, train_y, test_y = train_test_split(X, y, random_state=42)
model = LogisticRegression(random_state=42)
model.fit(train_X, train_y)
pred_y = model.predict(test_X)
print(np.where(pred_y == 1, True, False))
print(np.where(pred_y == 1))
# In[162]:
XI = np.linspace(-10, 10) #-10~10사이에 난수 생성, 이 때 간격은 거의 동일하다.
Y = -model.coef_[0][0] / model.coef_[0][
1] * XI - model.intercept_ / model.coef_[0][1]
plt.plot(XI, Y)
plt.xlim(min(X[:, 0]) - 0.5, max(X[:, 0]) + 0.5)
plt.ylim(min(X[:, 1]) - 0.5, max(X[:, 1]) + 0.5)
plt.axes().set_aspect('equal', 'datalim')
# In[171]:
model.coef_[0, 0]
model.intercept_
# In[ ]:
def feature_visualization3(method, dimension, fold):
color_choice = ['b', 'g', 'r', 'yellow']
x_numpy = np.loadtxt('experiments/features.txt')
x_numpy = x_numpy[54 * fold:54 * (fold + 1)]
my_labels = np.loadtxt('experiments/features_labels.txt')
my_labels = my_labels[54 * fold:54 * (fold + 1)]
print('Size of labels {}'.format(my_labels.shape))
# print('labels {}'.format(my_labels))
died_index = np.where(my_labels == 0)
survived_index = np.where(my_labels == 1)
if method == 'tsne':
module = manifold.TSNE(n_components=dimension,
init='pca',
random_state=100,
early_exaggeration=2000,
method='exact')
elif method == 'PCA':
module = PCA(n_components=10)
x_reduced = module.fit_transform(x_numpy)
died_vector = x_reduced[died_index]
survived_vector = x_reduced[survived_index]
print(x_reduced)
# plt.subplot(2, 5, i+1, figsize=(15, 15))
plt.figure(figsize=(5, 5))
if dimension == 2:
plt.plot(died_vector[:, 0],
died_vector[:, 1],
'o',
color='r',
alpha=0.5,
label='Deceased')
plt.plot(survived_vector[:, 0],
survived_vector[:, 1],
'o',
color='g',
alpha=0.5,
label='Survived')
# axs[i//5, i%5].plot(died_vector[:, 0], died_vector[:, 1], 'o', color='r', alpha=0.5, label='Died')
# axs[i//5, i%5].plot(survived_vector[:, 0], survived_vector[:, 1], 'o', color='g', alpha=0.5, label='Survived')
# axs[i // 5, i % 5].axis('off')
# axs[i // 5, i % 5].set_xticks([])
# axs[i // 5, i % 5].set_yticks([])
# plt.plot(x_reduced[i, 0], x_reduced[i, 1], 'o', color=color_choice[color_index])
plt.xticks([])
plt.yticks([])
elif dimension == 3:
ax = plt.axes(projection='3d')
ax.scatter(died_vector[:, 0],
died_vector[:, 1],
died_vector[:, 2],
color='r',
label='Died')
ax.scatter(survived_vector[:, 0],
survived_vector[:, 1],
survived_vector[:, 2],
color='g',
label='Survived')
ax.set_xlabel('X Label')
ax.set_ylabel('Y Label')
ax.set_zlabel('Z Label')
# plt.title('t-SNE Visualization of Patch-34 Feature Vectors')
plt.legend(loc="lower right")
plt.savefig('experiments/tsne/{}_{}_{}.pdf'.format(method, dimension,
fold))
plt.savefig('experiments/tsne/{}_{}_{}.jpg'.format(method, dimension,
fold))
plt.show()
plt.figure(figsize=(8, 6))
plt.pie(values,
labels=pollutants,
explode=explode,
autopct='%1.1f%%',
shadow=True)
plt.title('Air pollutants and their probable amount in atmosphere [India]')
plt.axis('equal')
plt.show()
# # showing INDIA AQI on world map using cartopy
# In[82]:
import cartopy.crs as ccrs
# In[83]:
geo = data['city']['geo']
fig = plt.figure(figsize=(12, 10))
ax = plt.axes(projection=ccrs.PlateCarree())
ax.stock_img()
plt.scatter(geo[1], geo[0], color='blue')
plt.text(geo[1] + 3, geo[0] - 2, f'{name} AQI \n {aqi}', color='red')
plt.show()
velocities = np.array(list(zip(np.sin(angles), np.cos(angles))))
def apply_boundary_conditions(self):
"""apply boundary conditions"""
edge_buffer_radius = 2.0
for position in self.positions:
x = position[0]
y = position[1]
if x > width + edge_buffer_radius:
x = -edge_buffer_radius
elif x < -edge_buffer_radius:
x = width + edge_buffer_radius
if y > height + edge_buffer_radius:
y = -edge_buffer_radius
elif y < -edge_buffer_radius:
y = height + edge_buffer_radius
position[0] = x
position[1] = y
figure = plt.figure()
axes = plt.axes(xlim=(0, width), ylim=(0, height))
bird_bodies, = axes.plot([], [], markersize=10, c='k', marker='o', ls='None')
#TODO Pick up here from page 76, code line 2 beak == bird_beak
skeleton = np.array(skeleton) - 1
for i in skeleton:
if pose_3d[i[0]][2] != 0 and pose_3d[i[1]][2]:
x = np.array([pose_3d[i[0]][0], pose_3d[i[1]][0]])
y = np.array([pose_3d[i[0]][1], pose_3d[i[1]][1]])
z = np.array([pose_3d[i[0]][2], pose_3d[i[1]][2]])
ax.plot3D(x, y, z, 'red')
def add_point(pose_3d, ax):
x, y, z = pose_3d[:, 0], pose_3d[:, 1], pose_3d[:, 2]
ax.scatter3D(x, y, z, cmap='Greens')
import matplotlib.pyplot as plt
fig = plt.figure()
ax = plt.axes(projection='3d')
plt.ion()
i = 0
for pose_3d in pose_3d_all:
add_line(pose_3d, ax)
add_point(pose_3d, ax)
square_x = np.array([-2000, 1000])
square_y = np.array([-1000, 2000])
square_z = np.array([500, 3500])
ax.scatter3D(square_x, square_y, square_z, color='whitesmoke')
plt.draw()
ax.set(aspect=1)
#----------------- drawing box -------------------------------#
xlim = ax.get_xlim()
ylim = ax.get_ylim()
box_x = [xlim[0], xlim[1], xlim[1], xlim[0], xlim[0]]
box_y = [ylim[0], ylim[0], ylim[1], ylim[1], ylim[0]]
ax.add_patch(
patches.Rectangle((box_x[0], box_y[0]),
xlim[1] - xlim[0],
ylim[1] - ylim[0],
fill=True,
facecolor='white',
clip_on=False,
zorder=0.8))
ax.plot(box_x, box_y, color='blue', linewidth=5.0)
plt.axes(ax)
# 0 1
# ax = plt.subplot()
# dy = (inverse((1, 0)) - inverse((1, -0.1)))[1]
#xy = transform((j+0.8, energydif[j]))
# print(dy, xy)
# ax = plt.axes([xy[0], xy[1]-(dy)/2.6, 0.170, 0.170])
cell = atoms.get_cell()
img = atoms.copy()
plot_conf(ax, img, colorlenth, rot=True)
ax.set_xlim([centreofmass[0] - 11.5, centreofmass[0] + 11.5])
ax.set_ylim([centreofmass[1] - 10.5, centreofmass[1] + 11.0])
ax.set_yticks([])
ax.set_xticks([])
ax.set(aspect=1)
def plotFeatureArrays(featureVecs,
array_shape,
n_figs=10,
tiled=True,
tiles_shape=(2, 5),
tile_psn=(.1, .125, .4, .4),
xlims=None,
ylims=None,
titles=None,
xlabel=None,
ylabel=None,
noise_floor=None,
extent=None,
origin=None,
colorbar=True):
n_samples = featureVecs.shape[0]
if tiled:
sample_idxs = np.array(random.sample(range(n_samples)),
np.prod(tiles_shape)).reshape(tiles_shape)
plt.figure()
for r in range(tiles_shape[0]):
for c in range(tiles_shape[1]):
idx = sample_idxs[r, c]
arr = featureVecs[idx, :].reshape(array_shape)
# set signal limits
maxSig = arr.max()
if noise_floor is not None:
minSig = magSig - noise_floor
arr[arr < minSig] = minSig
minSig = arr.min()
left = tile_psn[0] * (c + 1) + tile_psn[2] * c
bottom = tile_psn[1] * (r + 1) + tile_psn[3] * r
plt.axes((left, bottom, tile_psn[2], tile_psn[3]))
plt.imshow(arr,
extent=extent,
aspect='auto',
interpolation='nearest',
origin=origin,
cmap='binary',
vmin=minSig,
vmax=maxSig)
if colorbar:
plt.colorbar()
if xlims is not None:
plt.xlim(xlims)
if ylims is not None:
plt.ylim(ylims)
if xlabel is not None and r == 0:
plt.xlabel(xlabel)
else:
plt.xticks([])
if ylabel is not None and c == 0:
plt.ylabel(ylabel)
else:
plt.yticks([])
if titles is not None:
plt.title(title[idx])
plt.show()
else:
sample_idxs = np.array(random.sample(range(n_samples)), n_figs)
for idx in sample_idxs:
arr = featureVecs[idx, :].reshape(array_shape)
# set signal limits
maxSig = arr.max()
if noise_floor is not None:
minSig = magSig - noise_floor
arr[arr < minSig] = minSig
minSig = arr.min()
plt.figure()
plt.axes(tile_psn)
plt.imshow(arr,
extent=extent,
aspect='auto',
interpolation='nearest',
origin=origin,
cmap='binary',
vmin=minSig,
vmax=maxSig)
if colorbar:
plt.colorbar()
if xlims is not None:
plt.xlim(xlims)
if ylims is not None:
plt.ylim(ylims)
if xlabel is not None and r == 0:
plt.xlabel(xlabel)
else:
plt.xticks([])
if ylabel is not None and c == 0:
plt.ylabel(ylabel)
else:
plt.yticks([])
if titles is not None:
plt.title(title[idx])
plt.show()
ages = user_fields.map(lambda x: int(x[1])).collect() hist(ages, bins=20, color='lightblue', normed=True) fig = plt.pyplot.gcf() fig.set_size_inches(16, 10) count_by_occupation = user_fields.map(lambda fields: (fields[3], 1)).reduceByKey(lambda x, y: x + y).collect() x_axis1 = np.array([c[0] for c in count_by_occupation]) y_axis1 = np.array([c[1] for c in count_by_occupation]) x_axis = x_axis1[np.argsort(y_axis1)] y_axis = y_axis1[np.argsort(y_axis1)] pos = np.arange(len(x_axis)) width = 1.0 ax = plt.axes() ax.set_xticks(pos + (width / 2)) ax.set_xticklabels(x_axis) plt.bar(pos, y_axis, width, color='lightblue') plt.xticks(rotation=30) fig = plt.pyplot.gcf() fig.set_size_inches(16, 10) count_by_occupation2 = user_fields.map(lambda fields: fields[3]).countByValue() print "Map-reduce approach:" print dict(count_by_occupation2) print "" print "countByValue approach:" print dict(count_by_occupation) '''
#Define v_jhat_t - as v[1](y) multiplied by transformed vector jhat
v_jhat_t = v[1] * jhat_t
#TODO 3.:Define transformed vector v(v_t) as
#vector v_ihat_t added to vector v_jhat_t
v_t = None
#Plot that graphically shows vector v (color = 'skyblue') can be transformed
#into transformed vector v(v_trfm - color ='b') by adding v[0]*transformed
#vector ihat to v[0]*transformed vector jhat
#Creates axes of plot referenced 'ax'
ax = plt.axes()
#Plots red dot at origin (0,0)
ax.plot(0, 0, 'or')
#Plots vector v_ihat as dotted green arrow starting at origin 0,0
ax.arrow(0,
0 * v_ihat,
color='g',
linestyle='dotted',
linewidth=2.5,
head_width=0.30,
head_length=0.35)
#Plots vector v_jhat_t as dotted red arrow starting at origin defined by v_ihat
ax.arrow(v_ihat - t[0],
def create_surface(self): #create initial surface to 'erode'
veclength = self.length * self.width #create vector with enough numbers to fill a length*width matrix
mean, sd = 0, self.height #height is standard deviation around mean height of zero
heightlist = np.random.normal(
mean, sd, veclength
) #samples the proper number of random heights, around a mean of 0, with sd defined by 'height'
heightlist = np.around(
heightlist,
2) #rounds the heights to the nearest two decimal places
heightlist = heightlist.reshape(
(self.width, self.length)
) #turns the heights into a matrix with the proper dimensions for length and width
print(heightlist)
xlist = [] #creates empty list to hold x values (correspond to width)
ylist = [] #creates empty list to hold y values (correspond to length)
zlist = [] #creates empty list to hold z values (correspond to height)
for y in range(self.length): #for each y value from 0 to the length...
for x in range(
self.width): #for each x value from 0 to the width...
xlist.append(x) #add that x value in to a list of x values
ylist.append(y) #add that y value to a list of y values
z = heightlist[
x,
y] #find the z value (already in the matrix) for that [x,y] pair
print('Z: ', str(z))
zlist.append(z) #add that z value to the list of z values
print('XLIST: ', xlist)
print('LENXLST', str(len(xlist)))
print('YLIST: ', ylist)
print('LENYLIST: ', str(len(ylist)))
print('ZLIST: ', zlist)
print('LENZLIST: ', str(len(zlist)))
#TODO turn length and width into meshgrid
lengthlist, widthlist = np.arange(self.length), np.arange(
self.width
) #create arrays containing all possible numbers in length and width values
meshlength, meshwidth = np.meshgrid(
lengthlist,
widthlist) #create meshgrid for those length and width arrays
print('MESHLENGTH:')
print(meshlength)
print('MESHWIDTH:')
print(meshwidth)
#TODO select random points out of length, width and height
num_sampled = math.floor(
len(xlist) / 4
) #use around 1/4 of the points in the matrix as sample points for extrapolation
sample_points = random.sample(
range(len(xlist)), num_sampled
) #knowing the number of points to be sampled, chose the sample point indices
print('SAMPLE POINTS: ', sample_points)
xlist, ylist, zlist = np.asarray(xlist), np.asarray(ylist), np.asarray(
zlist) #turn the x,y, and z lists into np arrays
xsamples, ysamples, zsamples = xlist[[sample_points]], ylist[[
sample_points
]], zlist[[
sample_points
]] #choose the sample points from the arrays based on predetermined indices (same for each)
print('XSAMPLES: ', xsamples)
print('XYZsamples done')
print('XSAMPLES TYPE: ', type(xsamples))
print('YSAMPLES TYPE: ', type(ysamples))
print('ZSAMPLES TYPE: ', type(zsamples))
gridz0 = griddata(
(xsamples, ysamples), zsamples,
(meshwidth, meshlength
)) #apply griddata to extrapolate the rest of the height values
print('GRIDZ0: ', gridz0)
fig = plt.figure()
ax = plt.axes(projection='3d') #create 3D projection
ax.plot_surface(
meshwidth, meshlength, gridz0
) #plot it using meshwidth as X, meshlength as Y, and gridz0 as Z
plt.show()
def feature_visualization2(method, dimension):
color_choice = ['b', 'g', 'r', 'yellow']
x1 = np.load('./ChaData/feature1.npy')
print(x1.shape)
# y1 = np.ones_like(x1)
x2 = np.load('./MyData/feature1.npy')
print(x2.shape)
# y2 = np.zeros_like(x2)
# x2 = x2[100:x2.shape[0]]
x_numpy = np.zeros((x1.shape[0] + x2.shape[0], 4096))
my_labels = np.zeros((x1.shape[0] + x2.shape[0]))
x_numpy[0:x1.shape[0]] = x1
x_numpy[x1.shape[0]:x1.shape[0] + x2.shape[0]] = x2
my_labels[0:x1.shape[0]] = 1
my_labels[x1.shape[0]:x1.shape[0] + x2.shape[0]] = 0
# x_numpy = np.load('./ChaData/feature.npy')
# my_labels = np.load('./ChaData/label.npy')
print('Size of data {}'.format(x_numpy.shape))
print('Size of labels {}'.format(my_labels.shape))
died_index = np.where(my_labels == 0)
survived_index = np.where(my_labels == 1)
if method == 'tsne':
module = manifold.TSNE(n_components=dimension,
init='pca',
random_state=100,
early_exaggeration=20,
method='exact')
elif method == 'PCA':
module = PCA(n_components=10)
x_reduced = module.fit_transform(x_numpy)
print(x_reduced)
died_vector = x_reduced[died_index]
survived_vector = x_reduced[survived_index]
plt.figure()
if dimension == 2:
plt.plot(died_vector[:, 0],
died_vector[:, 1],
'ro',
label='Target Domain') #alpha=0.4
plt.plot(survived_vector[:, 0],
survived_vector[:, 1],
'b^',
label='Source Domain') #, alpha=0.4
# plt.plot(x_reduced[i, 0], x_reduced[i, 1], 'o', color=color_choice[color_index])
plt.xticks()
plt.yticks()
# fig = plt.gcf()
# dna1name='tsne'+'.png'
# fig.savefig(dna1name,dpi=100)
elif dimension == 3:
ax = plt.axes(projection='3d')
ax.scatter(died_vector[:, 0],
died_vector[:, 1],
died_vector[:, 2],
color='r',
label='Target Domain')
ax.scatter(survived_vector[:, 0],
survived_vector[:, 1],
survived_vector[:, 2],
color='g',
label='Source Domain')
ax.set_xlabel('X Label')
ax.set_ylabel('Y Label')
ax.set_zlabel('Z Label')
# plt.title('t-SNE Visualization of Source And Target Domain Feature Vectors')
plt.legend(loc="upper right")
plt.savefig('tsne1.png', dpi=300)
plt.show()
return x_reduced
u_var = one_sig_upper - tqmed
l_var = tqmed - one_sig_lower
print("{0} stack, bin size = {1}, tq =".format(stack, bin_size), tqmed,
'+', u_var, '-', l_var)
#tqmed = 5.9
cov_interp = interpol.interp1d(tqs, mod_covars, axis=0)
#flux_covar = cov_interp(8.7)
flux_covar = cov_interp(tqmed)
fig = plt.figure()
ax = plt.axes(None, label=str(k))
corr = flux_covar / np.sqrt(
np.outer(np.diag(flux_covar), np.diag(flux_covar)))
im = ax.pcolormesh(qstack.wave_grid[:last_in + 1],
qstack.wave_grid[:last_in + 1], corr)
plt.ylim(qstack.wave_grid[-2], qstack.wave_grid[0])
cbar = fig.colorbar(im, ax=ax)
cbar.set_label('Correlation')
#plt.title("Continuum Normalized Flux Covariance Matrix Across Wavelength Bins")
plt.xlabel("Wavelength Bin (Angstroms)")
plt.ylabel("Wavelength Bin (Angstroms)")
plt.savefig('model/corr_{1}_tq{0:.2f}.pdf'.format(tqmed, bin_size))
plt.figure()
std = np.sqrt(np.diagonal(flux_covar))
mean_mod = model(waves, tqmed)
Qs=abs((masses[2]-masses[1])/(masses[1]-masses[0]))
Qs_target=abs((drs[2]**order-drs[1]**order)/(drs[1]**order-drs[0]**order))
print 'Qs =',Qs,'Qs target =',Qs_target
# PLOT RESULTS ------------------------------------------------------------------
ls=['-.','--','-'] #line styles
#plot masses, pressures, density for best resolution run
r=arange(0.,drs[-1]*isurf/1e5,drs[-1]/1e5) # in km
rect=[0.1,0.1,0.8,0.8]
print r[-1], tovs[-1][isurf,3]/msun
a1=axes(rect)
a1.yaxis.tick_left()
plot(r,tovs[-1][:isurf,1]/pressc,'r', # plot pressure
r,tovs[-1][:isurf,0]/rhoc,'g') # and density
ylim([0,1.05])
xscale('log')
ylabel('Pressure (P_c), Density (rho_c)')
xlabel('Radius (km)')
a2=axes(rect,frameon=False)
a2.yaxis.tick_right()
plot([0],[0],'r',[0],[0],'g',r,tovs[-1][:isurf,3]/msun,'b') # plot mass
a2.yaxis.set_label_position('right')
ylabel('Mass (Msun)')
a2.set_xticks([])
