def plotTradeVsNews(tickName):
path2 = "resultsMKII"
frame = getNewsNTradingVol(tick_Name,path2)
newsBuz = []
tradingVol = []
newsVol = []
for i in range(len(frame['tradingVol'])):
newsBuz.append(frame['NewsBuz'].values[i])
tradingVol.append(np.log(frame['tradingVol'].values[i]))
newsVol.append(np.log(frame['NewsVol'].values[i]))
sns.set(style="ticks")
x = np.array(newsBuz)
y = np.array(tradingVol)
ax = sns.jointplot(x,y,kind="hex",stat_func=kendalltau,color="#4CB391")
ax.set_axis_labels(xlabel= "News Buz",ylabel="Trading Volume")
g = sns.jointplot(x, y, kind="kde", size=7, space=0)
g.set_axis_labels(xlabel= "News Buz",ylabel="Trading Volume")
x = np.array(newsVol)
ay = sns.jointplot(x,y,kind="hex",stat_func=kendalltau,color="#4CB391")
ay.set_axis_labels(xlabel= "News Volume",ylabel="Trading Volume")
h = sns.jointplot(x, y, kind="kde", size=7, space=0)
h.set_axis_labels(xlabel= "News Volume",ylabel="Trading Volume")
sns.plt.show()
# sns.plt.subplot(2,1,1)#41B3D3
# a1 = sns.regplot(x="NewsBuz", y="tradingVol", data=frame,ci=None,fit_reg=False,color="#1dad9b")
# a1.set_ylim([0,4e8])
# sns.plt.subplot(2,1,2)
#
# a2 = sns.regplot(x="NewsVol", y="tradingVol", data=frame,ci=None,fit_reg=False,color="#41B3D3")
# a2.set_ylim([0,4e8])
sns.plt.show()
def doplot(self, name):
"""
Do some plots
"""
self.trace = pickle.load( open( name, "rb" ) )
var = np.vstack([self.trace['muCB'][:,0], self.trace['muCB'][:,1], self.trace['sdCB'][:,0], self.trace['sdCB'][:,1]]).T
corner.corner(var, labels=['$\mu_C$', '$\mu_B$', '$\sigma_C$','$\sigma_B$'], show_titles=True)
pl.show()
# pl.savefig('{0}.png'.format(name))
# Just get the first N samples. We shuffle the
# arrays and get the subsamples
C = self.trace['CB'][:,:,0]
np.random.shuffle(C)
C_slice = C[0:200,:].flatten()
B = self.trace['CB'][:,:,1]
np.random.shuffle(B)
B_slice = B[0:200,:].flatten()
# First option
pl.plot(B_slice, C_slice, '.', alpha=0.002)
pl.show()
# KDE joint plot
sns.jointplot(C_slice, B_slice, kind='kde')
pl.show()
def seaborn_join():
data = np.random.multivariate_normal([0, 0], [[5, 2], [2, 2]], size=2000)
data = pd.DataFrame(data, columns=['x', 'y'])
with sns.axes_style('white'):
sns.jointplot("x", "y", data, kind='hex')
plt.show()
def histogram(self,x=None, y=None, l=None, t=None, **kwargs):
"""
this is a short-cut for creating many possible histograms, at a
specified beamline location l, or specified time t.
- if x and y are not input, then it creates a full joint-scatterplot
for each pair of variables (7 variables total: x,y,z, vx, vy, vz, t)
- if x is input, it creates a 1d histogram with respect to that parameter
- if x and y are input, creates a 2d histogram with respect to those parameters
"""
table = self.to_dataframe(l=l, t=t, latex=True)
if x is None and y is None:
g = sns.pairplot(table, **kwargs)
for ax in g.axes.flat:
_ = plt.setp( ax.xaxis.get_majorticklabels(), rotation=90)
return
if x is not None and y is None:
x = self._reformat_label(x)
sns.distplot(table[x], **kwargs)
plt.xlabel(x)
return
if x is not None and y is not None:
x = self._reformat_label(x)
y = self._reformat_label(y)
sns.jointplot(x=x, y=y, data=table, **kwargs);
return
def make_scatter_plot(frame, name, **kwargs):
"""
Makes a scatter plot of column name in frame.
"""
column_x = frame[name]
if name == 'deltam31': column_x*=100.0
params = []
exclude = set(['hypo','llh','mctrue'])
params = list(set(frame.columns).difference(exclude))
figs = []
# Plot correlation scatter plot for all other systematics
for p in params:
if p == name: continue
column_y = frame[p]
if p == 'deltam31': column_y*=100.0
if 'theta' in p: column_y = np.rad2deg(column_y)
with sns.axes_style("whitegrid"):
sns.jointplot(column_x, column_y, size=8, color='b',
**kwargs)
plt.tight_layout()
figs.append(plt.gcf())
return figs
def plotBonusvsSalary(df):
sns.jointplot(x="bonus", y="salary", data=df)
fig = plt.gcf()
fig.set_size_inches(18.5, 10.5)
fig.savefig('bonusVSsalary.png', dpi=100)
#plt.savefig('bonusVSsalary.png')
plt.show()
def plot_seaborn( self ): # https://stanford.edu/~mwaskom/software/seaborn/tutorial/distributions.html data = pd.read_csv( 'movement.csv' ).as_matrix() # 1/2 3/4 5/6 7/8 x_column = 3 y_column = 4 limit = 100 data = data[ ( data[:,0] == 0) & ( data[:,x_column] > -limit ) & ( data[:,x_column] < limit ) & ( data[:,y_column] > -limit ) & ( data[:,y_column] < limit ) ] x = data[:,x_column] y = data[:,y_column] with sns.axes_style( 'white' ): sns.jointplot( x=x, y=y, kind='kde' ) # scatter, reg, resid, hex, kde sns.plt.show()
def skill_vs_speed(prediction_mode, time_model, data):
model = TimeCombiner(prediction_mode, time_model)
Evaluator(data, model).get_report(force_run=True)
students = data.get_students()
skills = prediction_mode.get_skills(students)
fastness = time_model.get_skills(students)
sns.jointplot(pd.Series(skills), pd.Series(fastness), kind='kde', space=0).set_axis_labels("skill", "speed")
def show_graph(data):
""" Show time series graph of given data. """
height_list = sorted([[p[0], height(p[1:])] for p in data],
key=lambda x: x[0])
df = pd.DataFrame(height_list)
df.columns = ["time","height"]
seaborn.jointplot('time', 'height', data=df)
plt.show()
def sbratio(sampler):
chain = sampler.flatchain
chain[:,2]=np.abs(chain[:,2])
chain[:,4]=np.abs(chain[:,4])
dd = pd.DataFrame(data=chain,
columns=['theta','phi','scatter','badfrac','badsig','badmn'])
with sns.axes_style("white"):
sns.jointplot("theta", "phi", data, kind="kde");
def plot(self, samples, columns=None):
if(columns is None):
df = pd.DataFrame(samples, columns=["x", "y"])
sns.jointplot(x="x", y="y", data=df)
else:
df = pd.DataFrame(samples, columns=[columns[0], columns[1]])
# sns.jointplot(x=names[0], y=names[1], data=df, xlim=xlim, ylim=ylim)
sns.jointplot(x=columns[0], y=columns[1], data=df)
def plot_scatter_hist_sns(x, y):
#sns.set(color_codes=True)
#sns.set(style="darkgrid")
sns.set(style="ticks")
sns.jointplot(np.array(x), np.array(y), kind="hex", size=4, stat_func=None).set_axis_labels("$\phi$", "$\\theta$")
with PdfPages('plot4.pdf') as pdf:
pdf.savefig()
sns.plt.close()
def plot(data, total, title, width=800.0, unit='', dosort=True,
target=None, target2=None):
"""A HTML bar plot given a dictionary and max value."""
if len(data) > 30 and target is not None:
df = pandas.DataFrame(index=data)
df[title] = pandas.Series(data, index=df.index)
df[target.name] = target.ix[df.index]
if target2 is not None:
df[target2.name] = target2.ix[df.index]
if target.dtype == numpy.number:
if target2 is None:
seaborn.jointplot(target.name, title, data=df, kind='reg')
else:
seaborn.lmplot(target.name, title, data=df, hue=target2.name)
else: # X-axis is categorical
df.sort_values(by=target.name, inplace=True)
if target2 is None:
seaborn.barplot(target.name, title, data=df)
else:
seaborn.barplot(target.name, title, data=df, hue=target2.name)
fig = plt.gcf()
fig.autofmt_xdate()
# Convert to D3, SVG, javascript etc.
# import mpld3
# result = mpld3.fig_to_html(plt.gcf(), template_type='general',
# use_http=True)
# Convert to PNG
figfile = io.BytesIO()
plt.savefig(figfile, format='png')
result = '<div><img src="data:image/png;base64, %s"/></div>' % (
base64.b64encode(figfile.getvalue()).decode('utf8'))
plt.clf()
return result
result = ['<div class=barplot>',
('<text style="font-family: sans-serif; font-size: 16px; ">'
'%s</text>' % title)]
if target is not None:
data = OrderedDict([(key, data[key]) for key in
target.sort_values().index if key in data])
keys = {key.split('_')[0] if '_' in key else key[0] for key in data}
color = {}
if len(keys) <= 5:
color.update(zip(keys, range(1, 6)))
keys = list(data)
if dosort:
keys.sort(key=data.get, reverse=True)
for key in keys:
result.append('<br><div style="width:%dpx;" class=b%d></div>'
'<span>%s: %g %s</span>' % (
int(round(width * data[key] / total)) if data[key] else 0,
color.get(key.split('_')[0] if '_' in key else key[0], 1)
if data[key] else 0,
htmlescape(key), data[key], unit,))
result.append('</div>\n')
return '\n'.join(result)
def make_JointPlot(plot, region, data, backgrounds) :
sample_to_plot = []
if data.name == plot.sample : sample_to_plot.append(data)
if not len(sample_to_plot) :
for bk in backgrounds :
if bk.name == plot.sample : sample_to_plot.append(bk)
if len(sample_to_plot) == 0 or len(sample_to_plot) > 1 :
msg('ERROR make_JointPlot received %d samples to plot for plot with name %s'%(len(sample_to_plot), plot.name))
sys.exit()
# turn this tree into an array :)
sample_to_plot = sample_to_plot[0]
selection_ = '(' + region.tcut + ') * eventweight * ' + str(sample_to_plot.scale_factor)
tree_array = tree2rec(sample_to_plot.tree, branches=[plot.x_var, plot.y_var],
selection=selection_)
tree_array.dtype.names = (plot.x_var, plot.y_var)
x_arr = tree_array[plot.x_var]
y_arr = tree_array[plot.y_var]
sns.set(style="white")
# stats?
stat_func_ = None
if plot.stat_func == "kendalltau" :
from scipy.stats import kendalltau
stat_func_ = kendalltau
elif plot.stat_func == None :
from scipy.stats import pearsonr
stat_func_ = pearsonr
j_plot_grid = None
if plot.cmap == None or plot.cmap == "default" :
j_plot_grid = sns.jointplot(x_arr, y_arr, kind = plot.kind, stat_func=stat_func_, color = plot.color, linewidth = plot.line_width, ylim=[plot.y_range_min,plot.y_range_max], xlim=[plot.x_range_min,plot.x_range_max])
#j_plot_grid = sns.jointplot(x_arr, y_arr, kind = plot.kind, stat_func=stat_func_, color = plot.color, linewidth = plot.line_width, joint_kws={"n_levels":plot.n_levels, "shade":True}, ylim=[plot.y_range_min,plot.y_range_max], xlim=[plot.x_range_min,plot.x_range_max])
elif plot.cmap == "cubehelix" :
cmap_ = sns.cubehelix_palette(as_cmap=True, dark=0, light=1, reverse = True)
j_plot_grid = sns.jointplot(x_arr, y_arr, kind = plot.kind, stat_func=stat_func_, linewidth = plot.line_width, joint_kws={"cmap":cmap_, "n_levels":plot.n_levels, "shade":True}, ylim=[plot.y_range_min, plot.y_range_max], xlim=[plot.x_range_min,plot.x_range_max])
elif plot.cmap == "blues" :
j_plot_grid = sns.jointplot(x_arr, y_arr, kind = plot.kind, stat_func=stat_func_, linewidth = 1.0, joint_kws={"cmap":"Blues", "n_levels":plot.n_levels, "shade":True, "shade_lowest":False}, ylim=[plot.y_range_min, plot.y_range_max], xlim=[plot.x_range_min,plot.x_range_max])
else :
msg("cmap attribute of joint plot not yet added")
sys.exit()
j_plot_grid.fig.suptitle(plot.title)
j_plot_grid.fig.subplots_adjust(top=0.935)
j_plot_grid.set_axis_labels(plot.x_label, plot.y_label)
# save the plot to file
outname = plot.name + ".eps"
j_plot_grid.savefig(outname)
out = indir + "/plots/" + outdir
utils.mv_file_to_dir(outname, out, True)
fullname = out + "/" + outname
msg("%s saved to : %s"%(outname, os.path.abspath(fullname)))
def plot_approx_posterior(cov, means, index): mean = means[index] print mean.shape mean, cov = util.product_gaussians(mean, np.zeros(2), cov, np.identity(2)) data = np.random.multivariate_normal(mean, cov, 200) df = pd.DataFrame(data, columns=["x", "y"]) xlim = (mean[0] - 3*np.sqrt(cov[0][0]),mean[0] + 3*np.sqrt(cov[0][0])) ylim = (mean[1] - 3*np.sqrt(cov[1][1]),mean[1] + 3*np.sqrt(cov[1][1])) sns.jointplot(x="x", y="y", data=df, kind="kde", stat_func= None, xlim = xlim, ylim = ylim)
def plot_var(times, pitches, ends, var_n):
""" Show time series graph of variation [var_n]. """
# var_n: 0 to 30 (0: Aria)
n_data = filter(lambda x:(ends[var_n] < x[0] <= ends[var_n+1]),
zip(times, pitches))
# seaborn
df = pd.DataFrame(n_data)
df.columns = ["time","height"]
seaborn.jointplot('time', 'height', data=df)
plt.show()
def show(self):
Y = np.reshape(self._pr,(1,-1)).tolist()[0]
X = self._lams
df = pd.DataFrame({'x':X,'y':Y})
sns.jointplot(x='x',y='y',data=df)
Y = np.asarray(Y)
X = np.asarray(X)
mean = (X*Y).sum()
sns.plt.title('mean %f'%mean)
sns.plt.show()
def hist_2d(distribution, nsamples, **kwargs):
"""
Plots a 2d hexbinned histogram of distribution
"""
distr = distribution(ndims=2)
sampler = MarkovJumpHMC(distr.Xinit, distr.E, distr.dEdX, **kwargs)
samples = sampler.sample(nsamples)
with sns.axes_style("white"):
sns.jointplot(samples[0], samples[1], kind="kde", stat_func=None)
def pairwise_joint_plots(df, cols):
logging.debug('Plotting pairwise joint distributions')
cols = sorted(cols)
for colA, colB in [(a,b) for a in cols for b in cols if a < b]:
file = 'joint_{}_{}.png'.format(colA, colB)
logging.debug('joint plot: %s', file)
fig = plt.figure()
sns.jointplot(df[colA], df[colB], kind='hex')
plt.savefig(file)
plt.close()
def AnalyzeAllElectrodes():
"""From Jacek """
path = '/Users/ryszardcetnarski/Desktop/Nencki/Badanie_NFB/Dane/wszystkie_elektrody_jacek.csv'
db = pd.read_csv(path)
for band in ['theta', 'alpha','smr', 'beta1', 'beta2']:
db[band+'_po'] = db[band+ '_przed'] + db[band+'_roznica']
# fig = plt.figure()
# fig.suptitle(band)
# corr = fig.add_subplot(211)
# diff = fig.add_subplot(212)
sns.jointplot(band +'_przed', band+'_po', data=db, kind="reg")#, color="r", size=7)
# fig = plt.figure()
# fig.suptitle(band)
sns.jointplot(band +'_przed', band+'_roznica', data=db, kind="reg")#, color="r", size=7)
conditions_str = ['mixed_conditions' for i in range(0,len(db))]
conditions = [0 for i in range(0,len(db))]
GeneralModel( db[band+ '_przed'] , db[band+ '_po'] , band, conditions, conditions_str)
#corr.scatter(db[band +'_przed'], db[band+'_po'])
#diff.scatter(db[band +'_przed'], db[band+'_roznica'])
return db
#Kde using sklearn, returns object
# kde = KernelDensity(kernel='tophat', bandwidth = 3).fit(initial[:, np.newaxis])
# log_dens = kde.score_samples(x[:, np.newaxis])
#Plot sklearn kernel estimate
# kernel.plot(x, np.exp(log_dens), 'g')
#Plot original data histogram
#followUp = np.random.random_sample(100)
#followUp= np.random.normal(20,10, 100)
#followUp = np.random.normal(20,10, 100)#initial + np.random.normal(0,100,100)
#followUp = np.random.random_sample(100)#initial + np.random.normal(0,100,100)
#hist.hist(initial)
#initial = np.random.normal(20,10, 100)
#initial = np.random.random_sample(100)
#Add noise to each observation
#initial = #initial *0.95 + np.random.normal(100,100,100)
#Make a follow up by adding nosie second time to the same population
def plotCorrelation(frame):
# Plot correlation of each variable to visualize each dimension:
sns.jointplot("bedrooms","price",frame,size=8)
plt.tight_layout()
sns.jointplot("size","price",frame,size=8)
plt.tight_layout()
print("PAUSED...close figures to continue...")
plt.show()
return
def gauss_2d(nsamples=1000):
"""
Another simple test plot
1d gaussian sampled from each sampler visualized as a joint 2d gaussian
"""
gaussian = misc.distributions.TestGaussian(ndims=1)
control = Control(gaussian.Xinit, gaussian.E, gaussian.dEdX)
experimental = ContinuousTimeHMC(gaussian.Xinit, gaussian.E, gaussian.dEdX)
with sns.axes_style("white"):
sns.jointplot(control.sample(nsamples)[0], experimental.sample(nsamples)[0], kind="hex", stat_func=None)
def drawJointPlot(self, se1, se2):
"""
画线性相关图，表示序列1和序列2的相关性
:param self: 类变量本身
:param se1: 序列1
:param se2: 序列2
"""
sns.jointplot(se1, se2, kind='reg', color=self.linecolors[0])
# plt.title(self.title)
plt.legend()
plt.show()
def fixed_effects(data, labels):
corcoeff, p_val = pearsonr(data[labels[0]], data[labels[1]])
print "Pearson correlation between %s and %s across all donors is %g (two tailed p value = %g)"%(labels[0], labels[1], corcoeff, p_val)
grid = sns.jointplot(labels[0], labels[1], data, kind="hex")
sns.jointplot(labels[0], labels[1], data, kind="reg",
xlim=grid.ax_joint.get_xlim(),
ylim=grid.ax_joint.get_ylim())
plt.show()
return corcoeff, p_val
def covlen(args):
"""
%prog covlen covfile fastafile
Plot coverage vs length. `covfile` is two-column listing contig id and
depth of coverage.
"""
import numpy as np
import pandas as pd
import seaborn as sns
from jcvi.formats.base import DictFile
p = OptionParser(covlen.__doc__)
p.add_option("--maxsize", default=1000000, type="int", help="Max contig size")
p.add_option("--maxcov", default=100, type="int", help="Max contig size")
p.add_option("--color", default='m', help="Color of the data points")
p.add_option("--kind", default="scatter",
choices=("scatter", "reg", "resid", "kde", "hex"),
help="Kind of plot to draw")
opts, args, iopts = p.set_image_options(args, figsize="8x8")
if len(args) != 2:
sys.exit(not p.print_help())
covfile, fastafile = args
cov = DictFile(covfile, cast=float)
s = Sizes(fastafile)
data = []
maxsize, maxcov = opts.maxsize, opts.maxcov
for ctg, size in s.iter_sizes():
c = cov.get(ctg, 0)
if size > maxsize:
continue
if c > maxcov:
continue
data.append((size, c))
x, y = zip(*data)
x = np.array(x)
y = np.array(y)
logging.debug("X size {0}, Y size {1}".format(x.size, y.size))
df = pd.DataFrame()
xlab, ylab = "Length", "Coverage of depth (X)"
df[xlab] = x
df[ylab] = y
sns.jointplot(xlab, ylab, kind=opts.kind, data=df,
xlim=(0, maxsize), ylim=(0, maxcov),
stat_func=None, edgecolor="w", color=opts.color)
figname = covfile + ".pdf"
savefig(figname, dpi=iopts.dpi, iopts=iopts)
def main():
movie_raw_data = pd.read_csv('../input/movie_metadata.csv')
print movie_raw_data.head(3)
print movie_raw_data.isnull().sum()
print movie_raw_data.shape
movie_raw_data_dropna=movie_raw_data.dropna()
print movie_raw_data_dropna.shape
print movie_raw_data.dtypes
# movie_filterd_imdbscore=movie_raw_data['imdb_score'].loc
# movie_filterd_imdbscore=movie_raw_data.loc[movie_raw_data['imdb_score'].isin([2,3])]
movie_filterd_imdbscore_first=movie_raw_data.loc[movie_raw_data['imdb_score'] >5]
movie_filterd_imdbscore_from_raw=movie_raw_data.loc[movie_raw_data['imdb_score'] <8]
print movie_filterd_imdbscore_first.shape
movie_filterd_imdbscore_second=movie_filterd_imdbscore_first.loc[movie_raw_data['imdb_score'] <8]
print movie_filterd_imdbscore_second.shape
print movie_filterd_imdbscore_from_raw.shape
print '*********************************'
print movie_raw_data_dropna.head(3)
profit=(((movie_raw_data_dropna['gross'].values-movie_raw_data_dropna['budget'].values))/(movie_raw_data_dropna['gross'].values))*100
print profit
movie_raw_data_dropna.loc[:,'profit']=pd.Series(profit, movie_raw_data_dropna.index)
print movie_raw_data_dropna.shape
print movie_raw_data_dropna.head(3)
corr=movie_raw_data_dropna.corr()
print corr
f, ax = plt.subplots(figsize=(11, 9))
cmap = sns.diverging_palette(220, 10, as_cmap=True)
sns.heatmap(corr, cmap=cmap, vmax=1,
square=True,
linewidths=.5, cbar_kws={"shrink": .5}, ax=ax)
g = sns.jointplot(x="title_year", y="profit",kind='scatter',size=10,ylim = [0,110],xlim=[1980,2020],data=movie_raw_data_dropna)
h = sns.jointplot(x="imdb_score", y="profit",kind='reg',size=10,ylim = [0,110],data=movie_raw_data_dropna)
# j = sns.pairplot(movie_raw_data_dropna,hue='content_rating')
plt.show()
def sample_54_1():
"""
5.4 使用seaborn可视化数据
:return:
"""
sns.distplot(tsla_df['p_change'], bins=80)
plt.show()
sns.boxplot(x='date_week', y='p_change', data=tsla_df)
plt.show()
sns.jointplot(tsla_df['high'], tsla_df['low'])
plt.show()
def occupationAnalysis():
img = plt.imread("playground.jpg")
robot_position = readLog( "./csv/windfield_game1_green_withindex_position.csv")
data=np.zeros((nbcols, nbrows))
for robotp in robot_position:
robotp = robotp.split(',')
print(robotp[0])
px = int(float(robotp[1]) * nbcols)
py = int(float(robotp[2]) * nbrows)
data[px][py]+=1
robot_position = readLog( "./csv/windfield_game1_orange_withindex_position.csv")
for robotp in robot_position:
robotp = robotp.split(',')
print(robotp[0])
px = int(float(robotp[1]) * nbcols)
py = int(float(robotp[2]) * nbrows)
data[px][py]+=1
robot_position = readLog( "./csv/windfield_game1_blue_withindex_position.csv")
for robotp in robot_position:
robotp = robotp.split(',')
print(robotp[0])
px = int(float(robotp[1]) * nbcols)
py = int(float(robotp[2]) * nbrows)
data[px][py]+=1
fig, ax = plt.subplots()
#heatmap = ax.pcolor(data)
red_high = ((0., 0., 0.),
(.3, .5, 0.5),
(1., 1., 1.))
blue_middle = ((0., .2, .2),
(.3, .5, .5),
(.8, .2, .2),
(1., .1, .1))
green_none = ((0,0,0),(1,0,0))
cdict3 = {'red': red_high,
'green': green_none,
'blue': blue_middle,
'alpha': ((0.0, 0.0, 0.0),
(0.3, 0.5, 0.5),
(1.0, 1.0, 1.0))
}
#ax.scatter(x, y, label=str(i), color=color, alpha=0.5)
#dropout_high = LinearSegmentedColormap('Dropout', cdict3)
#plt.register_cmap(cmap = dropout_high)
sns.jointplot(x="x", y="y", data=data, kind="kde");
def performance_vs_coverage(db, output=None, max_values=250, **kwargs):
data = [
row for row in
db.execute(
"SELECT "
" performance AS performance, "
" coverage "
"FROM param_stats"
)
]
frame = pandas.DataFrame(data, columns=("Performance", "Legality"))
sns.jointplot("Legality", "Performance", data=frame,
xlim=(0, 1), ylim=(0, 1))
viz.finalise(output, **kwargs)
def show(self):
pos = np.argsort(self.pr)[0][-20:]
for k in pos:
print self.hypos[k],self.pr[0,k]
pos = np.argmax(self.pr)
print 'max',self.hypos[pos],'pr=',self.pr[0,pos]
X = []
for idx,hypo in enumerate(self.hypos):
N,f = hypo
X.append(idx)
Y = self.pr.tolist()[0]
df = pd.DataFrame({'x':X,'y':Y})
sns.jointplot(x='x',y='y',data=df)
sns.plt.show()
# In[25]:
# comapre with men and women that who have more target zero and who have not
fig, ax = plt.subplots(figsize=(10, 5))
sns.countplot(df['target'], hue=df['sex'], ax=ax)
plt.xlabel('target')
plt.ylabel('sex')
plt.xticks(rotation=50)
plt.show
# In[26]:
nums = ['age', 'sex', 'trestbps', 'chol', 'trestbps', 'target']
for i in nums:
plt.figure(figsize=(20, 10))
sns.jointplot(x=df[i], y=df['target'], kind='reg')
plt.xlabel(i)
plt.ylabel('resposne')
plt.grid()
plt.show()
# In[8]:
plt.bar(df['target'], df['age'], alpha=.5, width=0.8, label='chart')
plt.show()
# In[62]:
sns.catplot('sex', 'target', data=df, kind='box', hue='fbs')
# In[53]:
print("Minimum Cost: ${}".format(_min_cost))
print("Maximum Cost: ${}".format(_max_cost))
print("Mean Cost: ${}".format(_mean_cost))
print("Median Cost ${}".format(_median_cost))
print("Standard deviation of Cost: ${}".format(_stddev_cost))
_housedata['bedrooms'].value_counts().plot(kind='bar')
plt.title('Total number of Bedroom')
plt.xlabel('Bedrooms')
plt.ylabel('Count of Bedrooms')
plt.show()
#sns.despine
plt.figure(figsize=(10,10))
sns.jointplot(x=_housedata.lat.values, y=_housedata.long.values, size=10)
plt.ylabel('Longitude of House', fontsize=12)
plt.xlabel('Latitude of House', fontsize=12)
plt.show()
#plt1 = plt()
#sns.despine
plt.scatter(_housedata.price,_housedata.sqft_living)
plt.title("Price of House vs Square Feet of House")
plt.show()
plt.scatter(_housedata.price,_housedata.long)
plt.title("Price of House vs Location of the house area")
plt.show()
plt.scatter(_housedata.price,_housedata.lat)
names = ['variance','skewness','curtosis','entropy','class'])
data.head(3)
data.describe()
data.shape
data.isna().any()
data.dtypes
data['class'].unique()
sns.countplot(x='class', data= data)
sns.violinplot( y=data['curtosis'])
sns.violinplot( y=data['entropy'])
sns.violinplot( y=data['variance'])
sns.violinplot( y=data['skewness'])
p1=sns.kdeplot(data['curtosis'], shade=True, color="r")
p1=sns.kdeplot(data['variance'], shade=True, color="b")
sns.jointplot(x=data['curtosis'], y=data['entropy'], kind='hex', linewidth = 2)
sns.jointplot(x=data['skewness'], y=data['variance'], kind='hex', color = 'skyblue', linewidth = 2)
sns.jointplot(x=data['curtosis'], y=data['variance'], kind='hex', linewidth = 2)
X = data[['variance', 'skewness' ,'curtosis', 'entropy']]
y = data[['class']]
from sklearn.model_selection import train_test_split # Support Vector Machine
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 42)
from sklearn.svm import SVC
SVC()
svc = SVC()
# Creating a dictionary of parameters
parameters = {
def kde(self, n=0):
sns.jointplot(x=self.samples[:, n, 0],
y=self.samples[:, n, 1],
kind="kde")
def generate_plots(plot_type=""):
r"""
Generate plots studying the distribution of graphs in different splits with respect to the graph size (|V| and |E|)
:param plot_type: type of plot in {"histograms", "marginal_E", "marginal_V", "joint"}
"""
assert plot_type in {"histograms", "marginal_E", "marginal_V", "joint"}
split_names = ["test", "valid", "train"]
tot_n_nodes = []
tot_n_edges = []
for split_name in split_names:
d = ToulouseRoadNetworkDataset(split=split_name,
step=0.001,
max_prev_node=8)
dataloader = DataLoader(d,
batch_size=1,
shuffle=False,
collate_fn=custom_collate_fn)
n_nodes = []
n_edges = []
for datapoint in dataloader:
this_x_adj, this_x_coord, this_y_adj, this_y_coord, this_img, this_seq_len, this_id = datapoint
n_edges.append(int(this_y_adj.view(-1).sum().item()))
n_nodes.append(int(this_seq_len[0] - 2))
tot_n_edges += n_edges
tot_n_nodes += n_nodes
n_nodes = np.array(n_nodes)
n_edges = np.array(n_edges)
print(f"{split_name} min/mean/max len nodes", np.min(n_nodes),
np.mean(n_nodes), np.max(n_nodes))
print(f"{split_name} min/mean/max len edges", np.min(n_edges),
np.mean(n_edges), np.max(n_edges))
if plot_type == "histograms":
plt.hist(n_nodes, bins=np.max(n_nodes) - np.min(n_nodes) +
1) # arguments are passed to np.histogram
plt.title(f"Histogram of |V| for {split_name}")
plt.savefig(f"plots/histogram_|V|_{split_name}.png")
plt.clf()
plt.hist(n_edges, bins=np.max(n_edges) - np.min(n_edges) +
1) # arguments are passed to np.histogram
plt.title(f"Histogram of |E| for {split_name}")
plt.savefig(f"plots/histogram_|E|_{split_name}.png")
plt.clf()
elif plot_type == "marginal_V":
a = sns.kdeplot(n_nodes, bw=.5, shade=True, label=split_name)
elif plot_type == "marginal_E":
b = sns.kdeplot(n_edges, bw=.5, shade=True, label=split_name)
else:
sns_plot = sns.jointplot(np.log10(n_nodes),
np.log10(n_edges),
marginal_kws=dict(kernel="gau", bw=.02),
kind="kde",
bw=.05)
sns_plot.ax_joint.set_xlabel("log10 |V|", fontsize=15)
sns_plot.ax_joint.set_ylabel("log10 |E|", fontsize=15)
sns_plot.ax_marg_x.set_title(split_name, fontsize=20)
sns_plot.ax_joint.set_xlim(0.6, 1.2)
sns_plot.ax_joint.set_ylim(0.4, 1.2)
sns_plot.savefig(f"plots/joint_{split_name}.png")
tot_n_nodes = np.array(tot_n_nodes)
tot_n_edges = np.array(tot_n_edges)
print(f"min/mean/max len nodes", np.min(tot_n_nodes), np.mean(tot_n_nodes),
np.max(tot_n_nodes))
print(f"min/mean/max len edges\n", np.min(tot_n_edges),
np.mean(tot_n_edges), np.max(tot_n_edges))
if plot_type == "marginal_V":
a.set_xlabel("|V|")
a.set_ylabel("p(x)")
a.set_title("Distributions of |V|")
a.legend()
a.figure.savefig(f"plots/marginal_|V|.png")
a.figure.clf()
if plot_type == "marginal_E":
b.set_xlabel("|E|")
b.set_ylabel("p(x)")
b.set_title("Distributions of |E|")
b.legend()
b.figure.savefig(f"plots/marginal_|E|.png")
b.figure.clf()
print("Done!")
createFigure(
figure_data_without_zynex, 'EY_ROC', EARNINGS_YIELD, 'Return On Capital (%)',
'Earnings Yield (%)', 'ey_roc.png', 'lower right',
vscaling=1.2, hscaling=2)
createFigure(
figure_data, 'total_rank', 'EY_rank', 'Rank Return On Capital',
'Rank Earnings Yield', 'ey_roc_rank.png', 'upper right',
number_format='%d', vscaling=1.2, hscaling=2)
# Drop outliers
df_capped = df[df[EARNINGS_YIELD].between(
df[EARNINGS_YIELD].quantile(0.05), df[EARNINGS_YIELD].quantile(0.95))]
df_capped = df_capped[df_capped['ROC'].between(
df_capped['ROC'].quantile(0.05), df_capped['ROC'].quantile(0.95))]
# Save density plot
ax = sb.jointplot(EARNINGS_YIELD, 'ROC', data=df_capped, kind='kde', color="g")
ax.set_axis_labels('Earnings Yield (%)', 'Return On Capital (%)')
plt.tight_layout()
plt.savefig('density_plot.png', format='png')
plt.clf()
# Create industry histogram
ax = sb.countplot(x=SECTOR, data=figure_data, palette='Blues_d')
ax.set_xticklabels(ax.get_xticklabels(), rotation=45, ha='right')
ax.set_ylabel('Amount')
ax.set_xlabel('Industry')
plt.tight_layout()
plt.savefig('industry_histogram.png', format='png')
#%%
#Histogramme
seaborn.distplot(ordis.price)
#%%
# Boîte à moustaches
seaborn.factorplot("price", data=ordis, kind="box")
#%%
# violin
seaborn.factorplot("price", data=ordis, kind="violin")
#%%
# Lien entre price et var quanti (speed, hd)
seaborn.factorplot("speed", "price", data=ordis)
seaborn.jointplot("hd", "price", data=ordis, kind="reg")
#%%
# Lien entre price et var quali (ram, cd, premium, screen)
seaborn.factorplot("ram", "price", data=ordis, kind="box")
seaborn.factorplot("cd", "price", data=ordis, kind="box")
seaborn.factorplot("premium", "price", data=ordis, kind="box")
seaborn.factorplot("screen", "price", data=ordis, kind="box")
#%%
# price ~ speed et hd
t = pandas.crosstab(pandas.cut(ordis.hd, 6, precision=0),
ordis.speed,
values=ordis.price,
aggfunc=numpy.mean)
seaborn.heatmap(t, cmap="Blues", cbar_kws={'label': 'mean price'})
# In[18]: sns.pairplot(sub_task_summary_Output, hue='EV', palette='Set1') # In[20]: # SIMPLE LINE PLOT sub_task_summary_Output['EV'].plot(figsize=(20, 12)) # In[26]: # In[65]: plt.figure(figsize=(12, 8)) sns.jointplot(x='SPI', y='EV', data=sub_task_summary_Output, color='hotpink') sns.jointplot(x='CPI', y='EV', data=sub_task_summary_Output, color='red') sns.jointplot(x='EAC', y='EV', data=sub_task_summary_Output, color='blue') # # In[41]: # In[55]: # In[56]: # In[66]: # In[67]:
def heatscatter_sns(x, y, figsize=(8, 8)):
sns.set(rc={'figure.figsize': figsize})
sns.set(style="white", color_codes=True)
sns.jointplot(x=x, y=y, kind='kde', color="skyblue")
plt.figure(figsize=(10, 25))
sns.countplot(y='country', data=dataset, alpha=alpha)
plt.title('Data by country')
plt.show()
# Between Genders Male vs Female
plt.figure(figsize=(7, 7))
sex = sns.countplot(x='sex', data=dataset)
# Corelation between the Data
plt.figure(figsize=(16, 7))
cor = sns.heatmap(dataset.corr(), annot=True)
g = sns.jointplot(dataset.year,
dataset.suicides_no,
kind="kde",
color="#bfa9e0",
size=7)
plt.savefig('graph.png')
# Visualizing which age of people Suicide the most
plt.figure(figsize=(16, 7))
bar_age = sns.barplot(x='sex', y='suicides_no', hue='age', data=dataset)
# Visualizing which Generation of people Suicide the most
plt.figure(figsize=(16, 7))
bar_gen = sns.barplot(x='sex', y='suicides_no', hue='generation', data=dataset)
cat_accord_year = sns.catplot('sex',
'suicides_no',
hue='age',
df = DataFrame(iris.data,columns = iris.feature_names)
df['target'] = iris.target
print(df)
#数据可视化
import pandas as pd
from scipy import stats,integrate
import matplotlib.pyplot as plt
import seaborn as sns
sns.set(color_codes = True)
#数据分布可视化，直方图和密度函数
#distplot()函数默认绘出数据的直方图和密度函数
sns.distplot(df['petal length (cm)'],bins = 15)
#jointplot()函数同时绘制散点图和直方图
sns.jointplot(x = 'sepal length (cm)',y = 'sepal width (cm)',data = df,size =8)
#分组散点图
#用seaborn.FacetGrid标记不同的种类
sns.FacetGrid(df,hue = 'target',size =8).map(plt.scatter,'sepal length (cm)','sepal width (cm)').add_legend()
#六边形图
sns.axes_style('white')
sns.jointplot(x = 'sepal length (cm)',y = 'sepal width (cm)',data = df,kind = 'hex',color = 'r')
#二维核密度估计图
g = sns.jointplot(x = 'sepal length (cm)',y = 'sepal width (cm)',data = df,kind = 'kde',color = 'm')
#添加散点图
g.plot_joint(plt.scatter,c='w',s=30,linewidth=1,marker='+')
sns.distplot(bd['age'], kde=False, norm_hist=True, bins=10)
sns.distplot(bd['age'], hist=False)
sns.distplot(bd['age'], hist=False)
myimg = myplot.get_figure()
myimg.savefig('distplot.png')
sns.kdeplot(bd['age']) # other distribution plot, less used
sns.kdeplot(bd['age'], shade=True) # shade area
sns.kdeplot(bd['pdays'], shade=True)
myplot = sns.boxplot(y='age', data=bd)
myimg = myplot.get_figure()
myimg.savefig('boxplot.png')
myplot = sns.jointplot(x='age', y='balance', data=bd.iloc[:500, :])
myimg = myplot.get_figure() # not work in jointplot
myimg.savefig('jointplot.png')
myplot = sns.jointplot(x='age',
y='balance',
data=bd.iloc[:100, :],
kind='hex',
size=10)
# light colour less density,givenby hex
help(sns.jointplot)
sns.jointplot(x='age',
y='duration',
data=bd.iloc[:100, :],
kind='kde',
size=10)
myplot = sns.lmplot(x='age', y='balance', data=bd.iloc[1:10, :])
sns.distplot(data['x'])
sns.distplot(data['y'])
# In[9]:
for col in 'xy':
sns.kdeplot(data[col], shade=True)
# In[10]:
sns.kdeplot(data)
# In[12]:
with sns.axes_style('white'):
sns.jointplot("x", "y", data, kind='kde')
# In[13]:
with sns.axes_style('white'):
sns.jointplot("x", "y", data, kind='hex')
# In[14]:
sns.pairplot(data)
# In[20]:
import plotly.graph_objs as go
import numpy as np
x = np.random.randn(2000)
5, 5, 5, 5, 5, 10, 10, 10, 10, 10, 9, 9, 9, 9, 9, 9, 1, 11, 10, 10, 10, 10,
10, 10, 10, 8, 3, 7, 3, 2, 2, 2, 11, 7, 7, 11, 11, 9, 9, 8, 8, 8, 8, 7, 7,
7, 7, 7, 7, 7, 7, 7, 6, 12, 12, 12, 11, 11, 11, 9, 9, 9, 9, 9, 11, 11, 10,
1, 12, 12, 12, 3, 2, 12, 11, 11, 11, 11, 11, 11, 11, 10, 3, 11, 11, 2, 2,
1, 1, 1, 12, 12, 12, 12, 12, 12, 12, 6, 6
]
y = [
30, 29, 29, 24, 19, 11, 9, 8, 7, 3, 57, 54, 52, 34, 30, 29, 8, 1, 49, 44,
33, 31, 29, 29, 28, 27, 2, 6, 5, 52, 41, 36, 18, 27, 26, 46, 32, 35, 33,
15, 14, 10, 0, 51, 49, 44, 43, 28, 27, 26, 19, 16, 56, 21, 19, 16, 49, 43,
39, 25, 23, 22, 21, 13, 23, 1, 13, 17, 59, 55, 54, 10, 59, 1, 59, 57, 27,
25, 22, 21, 4, 49, 59, 31, 30, 5, 0, 8, 6, 0, 39, 37, 35, 31, 27, 25, 18,
11, 9
]
# print rs
# x = rs.gamma(12, size=60)
# y = 2 + rs.gamma(60,size=60)
# x = rs.gamma(2, size=1000)
# print 'y = '+ str(y)
graph = sns.jointplot(x, y, kind="hex", stat_func=kendalltau, color="#4CB391")
# x = np.random.normal(size=100)
# print 'x = '+ str(x)
# graph = sns.distplot(x);
sns.plt.savefig(__main__.__file__ + ".png")
# graph.pyplot.show()
sns.plt.show()
print("Kurtosis:")
print(data_set['T_MAX'].kurtosis())
## Graph T MAX / CO & O3
df = data_set.sort_values(['T_MAX', 'CO'], ascending=True)
plt.plot(df['T_MAX'], df['CO'])
plt.title("La concentración de CO frente a la temperatura máxima")
plt.show()
df = data_set.sort_values(['T_MAX', 'O3'], ascending=True)
plt.plot(df['T_MAX'], df['O3'])
plt.title("La concentración de Ozono frente a la temperatura máxima")
plt.show()
## Pairplot
sns.jointplot(data_set['T_MAX'], data_set['CO'], kind="reg")
plt.show()
plt.close()
sns.jointplot(data_set['T_MAX'], data_set['O3'], kind="reg")
plt.show()
plt.close()
## Correlation Matrix
data_set_corr = data_set
data_set_corr['Mes'] = data_set_corr['Mes'].map({
'ENE': 1,
'FEB': 2,
'MAR': 3,
'ABR': 4,
'MAY': 5,
'JUN': 6,
ax_histx = plt.axes(rect_histx) ax_histx.tick_params(direction='in', labelbottom=False) ax_histy = plt.axes(rect_histy) ax_histy.tick_params(direction='in', labelleft=False) # the scatter plot: ax_scatter.scatter(x, y) # now determine nice limits by hand: binwidth = 0.25 lim = np.ceil(np.abs([x, y]).max() / binwidth) * binwidth ax_scatter.set_xlim((-lim, lim)) ax_scatter.set_ylim((-lim, lim)) bins = np.arange(-lim, lim + binwidth, binwidth) ax_histx.hist(x, bins=bins) ax_histy.hist(y, bins=bins, orientation='horizontal') ax_histx.set_xlim(ax_scatter.get_xlim()) ax_histy.set_ylim(ax_scatter.get_ylim()) plt.show() # Seaborn version import numpy as np import seaborn as sns #sns.set(style="ticks") sns.jointplot(x, y) sns.jointplot(x, y, kind="hex", color="#4CB391")
#Visulization
matplotlib.rcdefaults()
plt.show(df.plot(kind = 'box'))
pd.options.display.mpl_style = 'default' # Sets the plotting display theme to ggplot2
df.plot(kind = 'box')
sns.boxplot(data=df,width=0.5)
sns.violinplot(df,width=3.5)
plt.show(sns.distplot(df.ix[:,2], rug = True, bins = 15))
with sns.axes_style("white"):
plt.show(sns.jointplot(df.ix[:,1],df.ix[:,2], kind = "kde"))
plt.show(sns.lmplot("Benguet","Ifugao",df))
#Creating custom function
def add_2int(x,y):
return x+y
print(add_2int(2,2))
# an algorithm example
def case(n=10,mu=3,sigma=np.sqrt(5),p=0.025,rep=100):
m=np.zeros((rep,4))
for i in range(rep):
norm = np.random.normal(loc = mu, scale = sigma, size = n)
xbar = np.mean(norm)
df.head()
import matplotlib.pyplot as plt
import seaborn as sns
df.groupby('title')['rating'].mean().sort_values(ascending=False).head()
df.groupby('title')['rating'].count().sort_values(ascending=False).head()
ratings = pd.DataFrame(df.groupby('title')['rating'].mean())
ratings['numRatings'] = pd.DataFrame(df.groupby('title')['rating'].count())
ratings.head()
ratings['numRatings'].hist(bins=100, figsize=(10, 6))
ratings['rating'].hist(bins=100, figsize=(10, 6))
sns.jointplot(x='rating', y='numRatings', data=ratings, alpha=0.6)
# as the number of ratings goes up, so does the average rating
moviemat = df.pivot_table(index='user_id', columns='title', values='rating')
moviemat.head()
ratings.sort_values('numRatings', ascending=False).head(10)
starwars_user_ratings = moviemat['Star Wars (1977)']
liarliar_user_ratings = moviemat['Liar Liar (1997)']
# This will show how people who have seen star wars rate other movies
similar_to_starwars = moviemat.corrwith(starwars_user_ratings)
similar_to_liarliar = moviemat.corrwith(liarliar_user_ratings)
import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd
#sns.residplot(x='age',y='fare',data=tips,color='indianred')
# Generate a green residual plot of the regression between 'hp' and 'mpg'
auto = pd.read_csv('auto.csv')
# Generate a joint plot of 'hp' and 'mpg'
sns.jointplot(x = 'hp', y = 'mpg', data = auto)
# Display the plot
plt.show()
axes1 = fig.add_axes([0.1, 0.1, 0.8, 0.8])
axes1.scatter(j_day, dw_solar_everyday, label='Observed dw_solar', color='red')
axes1.scatter(j_day, ghi_everyday, label='Clear Sky GHI', color='green')
axes1.set_xlabel('Days')
axes1.set_ylabel('Solar Irradiance (Watts /m^2)')
axes1.set_title('Solar Irradiance - Test Year 2009')
axes1.legend(loc='best')
fig.savefig('RNN Paper Results/Exp2_1/' + test_location + '/' + test_year +
'Figure 2.jpg',
bbox_inches='tight')
# In[525]:
sns.jointplot(x=dw_solar_everyday, y=ghi_everyday, kind='reg')
plt.xlabel('Observed global downwelling solar (Watts/m^2)')
plt.ylabel('Clear Sky GHI (Watts/m^2)')
plt.savefig('RNN Paper Results/Exp2_1/' + test_location + '/' + test_year +
'Figure 3',
bbox_inches='tight')
# ### making the Kt (clear sky index at time t) column by first removing rows with ghi==0
# In[526]:
if run_train:
# TRAIN dataset
df_train = df_train[df_train['ghi'] != 0]
df_train['Kt'] = df_train['dw_solar'] / df_train['ghi']
df_train.reset_index(inplace=True)
#mu = np.array([-0.5, -2.5])
size = 1000000 # at 10 million my RAM is overloaded
### If a vector X is normally distributed, then exp(X) is lognormally distributed with the same mean and variance
log_data = np.random.multivariate_normal(mu,cov, size=size)
level_data = np.exp(log_data)
k = level_data[:,1]
z = level_data[:,0]
lnk = log_data[:,1]
lnz = log_data[:,0]
### Plotting the joint density functions for levels and for logs
## First levels
sns.jointplot(k,z,kind="hex").set_axis_labels("Capital", "Productivity")
plt.show()
sns.jointplot(lnk,lnz,kind="hex").set_axis_labels("Log Capital", "Log Productivity")
plt.show()
'''
## Plotting the raw joint density of lognormal variables does not make much sense as in 10,000,000 observations there will be massive outliers
### I atempt to get rid of these outliers for plotting purposes
meank = np.mean(k)
sdk = np.std(k)
final_k = [x for x in k if (x > meank - 2 * sdk)]
final_k = [x for x in final_k if (x < meank + 2 * sdk)]
meanz = np.mean(z)
def viz_cont_cont(df, features, target):
for feature in features:
sns.jointplot(x=feature, y=target, data=df)
merged_df.popularity.plot.hist(bins=50, color='green')
# explore vote_average distribution
# appear to be almost normal distribution
merged_df.vote_average.plot.hist(bins=50, color='red')
# to fix popularity, we will remove vote_count under 10 to prevent bias
merged_df = merged_df[~(merged_df.vote_count < 10)]
# replot
merged_df.popularity.plot.hist(bins=50, color='blue',
alpha=0.5) # appear to be better
# plot scatter and find r2 for popularity versus domestic_gross columns
# before plot, we want to convert the scale into log10 and need to remove 0s
merged_df = merged_df[~(merged_df.domestic_gross == 0)]
merged_df = merged_df[~(merged_df.worldwide_gross == 0)]
merged_df.to_pickle('budget_popularity.pkl')
sns.jointplot(merged_df['popularity'],
np.log10(merged_df['domestic_gross']),
kind="reg",
stat_func=hf.r2)
sns.jointplot(merged_df['popularity'],
np.log10(merged_df['worldwide_gross']),
kind="reg",
stat_func=hf.r2)
sns.jointplot(merged_df['vote_average'],
np.log10(merged_df['domestic_gross']),
kind="reg",
stat_func=hf.r2)
sns.jointplot(merged_df['vote_average'],
np.log10(merged_df['worldwide_gross']),
kind="reg",
stat_func=hf.r2)
# popularity is R2 is 0.3 while vote_average is 0.051, we will use popularity as a metric to estimate gross income
# we will use popularity to estimate how well genres perform using tmdb data frame
for a, b in product(features, plottables):
msg('Making %s %s' % (a, b))
x = with_elo[a]
y = with_elo[b]
msg('type = %s' % x.dtype)
if x.dtype == 'object':
plt.figure()
x.value_counts().plot(kind='bar')
plt.savefig('/data/' + a + '_hist.png')
plt.close('all')
else:
try:
xlim = tuple(np.percentile(x, [1, 99]))
ylim = tuple(np.percentile(y, [1, 99]))
with sns.axes_style("white"):
sns.jointplot(x, y, kind="hex", xlim=xlim, ylim=ylim)
plt.savefig('/data/scatter_' + a + '_' + b + '.png')
plt.close('all')
except:
# sns.violinplot(x, y)
# plt.savefig('/data/' + a + '_' + b + '.png')
# plt.close()
plt.figure()
x.plot(kind='hist')
plt.savefig('/data/' + a + '_hist.png')
plt.close('all')
do_indivs = True
if do_indivs:
for a, b in product(features, plottables):
msg('Making %s %s' % (a, b))
file_out_figures = 'C:/Users/lalc/Documents/Old Documents folder/PhD/Meetings/July 2020/'
file = ['U','UN','N','SN']
limits = [[-np.inf,-.1], [-.1,-.01], [-.01,.01], [.01,.21], [.21,np.inf]]
limits = [[-np.inf,-.1], [-.1,-.01], [-.01,.01], [.01,.21]]
relind = L30min1.relscan>.25
j = -2
for i,l in enumerate(limits):
stabind = ((Ri1[:,j]>l[0]) & (Ri1[:,j]<l[1]))
cols = np.r_[['$L_{u_1,x_1}$', '$L_{u_1,x_2}$','$L_{v_1,x_1}$', '$L_{v_1,x_2}$','$L_{h,x_1}$', '$L_{h,x_2}$'], L30min1.columns [6:]]
L30min1.columns = cols
xlim = 5*200
ylim = 5*200
g = sns.jointplot(x ='$L_{h,x_1}$', y = '$L_{h,x_2}$', data=L30min1.loc[relind & stabind & ind1],
height = 8, kind="kde", cmap="jet", xlim = (0,xlim), ylim = (0,ylim),
color='k')#,cbar=True, cbar_kws={"format": formatter, "label": '$Density$'})
g.set_axis_labels('$L_{h,x_1}$', '$L_{h,x_2}$', fontsize = 24)
g.ax_joint.plot([0,xlim],[0,ylim],'--k', linewidth = 2)
g.ax_joint.plot(L30min1.loc[relind & stabind & ind1]['$L_{h,x_1}$'].values,L30min1.loc[relind & stabind & ind1]['$L_{h,x_2}$'].values,'o', color = 'k', alpha=.2)
g.ax_joint.text(100, 800,'$'+'%.2f' % l[0] +'<Ri_f<'+'%.2f' % l[1] +'$',fontsize=30,color='r')
plt.tight_layout()
plt.savefig(file_out_figures+file[i]+'_phase_1.png')
file = ['U','UN','N','SN','VS']
relind = L30min2.relscan>.25
for i,l in enumerate(limits):
stabind = ((Ri2[:,-2]>l[0]) & (Ri2[:,-2]<l[1]))
cols = np.r_[['$L_{u_1,x_1}$', '$L_{u_1,x_2}$','$L_{v_1,x_1}$', '$L_{v_1,x_2}$','$L_{h,x_1}$', '$L_{h,x_2}$'], L30min2.columns [6:]]
L30min2.columns = cols
# que pasa por los valores, lo desactivamos asi sns.distplot(tips['total_bill'],kde=False) plt.show() # podemos modificar la cantidad de bins que son la barras, # con el parametro bins solo pasando un int, hay que tener # cuidado con el tamaño del bin sns.distplot(tips['total_bill'],kde=False,bins=40) plt.show() # tenemos un metodos que nos compara dos columnas dentro de # un dataset sns.jointplot(x='total_bill',y='tip',data=tips) plt.show() # podemos graficar esto de varias maneras con el parametro # kind usando: hex, reg, kde # este otro metodo nos muestra una serie de graficas comparando # todas las columnas con todas, cuando se compara con si mismo, # muestra un histogram, y cuando es con otro, es un jointplot() sns.pairplot(tips) plt.show() # si queremos dividir la informacion de cada grafica por otras # columnas por ejemplo por sexo usamos el parametro hue, se le # pasa una columa categorial, no que tenga un valor por eso
Next compare the distributions of the positive and negative examples over a few features.
Good questions to ask yourself at this point are:
* Do these distributions make sense?
+ Yes. You've normalized the input and these are mostly concentrated in the +/- 2 range.
* Can you see the difference between the ditributions?
+ Yes the positive examples contain a much higher rate of extreme values.
-----------------------------------------------------------------------------------------
'''
pos_df = pd.DataFrame(train_features[ bool_train_labels], columns = train_df.columns)
neg_df = pd.DataFrame(train_features[~bool_train_labels], columns = train_df.columns)
sns.jointplot(
pos_df['V5'],
pos_df['V6'],
kind='hex',
xlim = (-5,5),
ylim = (-5,5)
)
plt.suptitle("Positive distribution")
sns.jointplot(
neg_df['V5'],
neg_df['V6'],
kind='hex',
xlim = (-5,5),
ylim = (-5,5)
)
_ = plt.suptitle("Negative distribution")
# Histogram
sns.distplot(a = iris_data['Petal Length (cm)'], kde=False)
# Kernel Density Estimate (kde)
# This is the smoothed histogram
# kde plot
sns.kdeplot(data=iris_data['Petal Length (cm)'], shade=True)
# We can create two-dimensional kde plot
sns.jointplot(x=iris_data['Petal Length (cm)'],
y=iris_data['Sepal Width (cm)'], kind='kde')
# Let split the data to understand difference btw species
iris_set_data = pd.read_csv('data/iris_setosa.csv', index_col="Id")
iris_ver_data = pd.read_csv('data/iris_versicolor.csv', index_col="Id")
iris_vir_data = pd.read_csv('data/iris_virginica.csv', index_col="Id")
def explore_global_plot(data, label='label', n_feats=50, id=None, task='classification'):
'''
:param data: DataFrame
:param label: label column name in the data
:param n_feats: the number of features be used to analysis.
:param task: regression or classification
:return:
'''
columns = data.columns.tolist()
columns.remove(label)
if id is not None:
if columns[id].duplicated().sum():
print('{} is duplicated !!!'.format(id))
columns.remove(id)
data.drop(id, axis=1, inplace=True)
numeric_features = [True if any([ptypes.is_integer_dtype(i),ptypes.is_int64_dtype(i),ptypes.is_float_dtype(i)]) else False for i in data[columns].dtypes]
numeric_names = [columns[i] for i, v in enumerate(numeric_features) if v]
category_names = list(set(columns) - set(numeric_names))
if task == 'classification':
if len(category_names):
# data distribution for each class
new_data = data.dropna(axis=0)
famd = prince.FAMD(
n_components=2,
n_iter=3,
copy=True,
check_input=True,
engine='auto',
random_state=42
)
famd = famd.fit(new_data[columns])
ax = famd.plot_row_coordinates(
new_data,
ax=None,
x_component=0,
y_component=1,
labels=new_data.index,
color_labels=['{}'.format(t) for t in new_data[label]],
ellipse_outline=False,
ellipse_fill=True,
show_points=True
)
plt.show()
else:
new_data = data.dropna(axis=0)
pca = PCA(n_components=2, random_state=seed)
X_pca = pca.fit_transform(new_data[columns])
sns.scatterplot(x=X_pca[:, 0], y=X_pca[:, 1], hue=label, data=new_data)
plt.show()
# sort features for correlation plot
sorted_feat_name = numeric_names
if len(numeric_names) > 6:
n_clusters = 3
new_data = data[[label] + numeric_names].dropna(axis=0)
new_data_feat = new_data[numeric_names]
new_data_stand = StandardScaler().fit_transform(new_data_feat)
kmean_init = KMeans(n_clusters=n_clusters, random_state=seed)
new_data_kmean=kmean_init.fit_transform(
new_data_stand.reshape(len(numeric_names), -1))
sorted_feat = sorted(zip(numeric_names, kmean_init.labels_), key=lambda x: x[1])
sorted_feat_name = [i[0] for i in sorted_feat]
# correlation plot for all features
sns.heatmap(data[[label] + sorted_feat_name + category_names].corr())
plt.show()
# outlier detection just for numeric features
outlier = data[numeric_names].apply(mad_based_outlier)
for i, column in enumerate(outlier.columns):
print('outlier:\n {}'.format(data[[column]][outlier.iloc[:, i]]))
# missing value pattern plot for all features
msno.matrix(data[columns[:n_feats]])
plt.show()
msno.bar(data[columns[:n_feats]])
plt.show()
miss_data = data[columns[:n_feats]].isnull().sum(axis=1)
miss_data = miss_data.to_frame()
miss_data.columns = ['number_of_missing_attributes']
miss_data.sort_values('number_of_missing_attributes', inplace=True)
miss_data['index'] = list(range(0, miss_data.shape[0]))
sns.jointplot(x="index", y="number_of_missing_attributes", data=miss_data)
plt.show()
def analyze_zN(z, outdir, vg, skip_umap=False, num_pcs=2, num_ksamples=20):
zdim = z.shape[1]
# Principal component analysis
log('Perfoming principal component analysis...')
pc, pca = analysis.run_pca(z)
log('Generating volumes...')
for i in range(num_pcs):
start, end = np.percentile(pc[:,i],(5,95))
z_pc = analysis.get_pc_traj(pca, z.shape[1], 10, i+1, start, end)
vg.gen_volumes(f'{outdir}/pc{i+1}', z_pc)
# kmeans clustering
log('K-means clustering...')
K = num_ksamples
kmeans_labels, centers = analysis.cluster_kmeans(z, K)
centers, centers_ind = analysis.get_nearest_point(z, centers)
if not os.path.exists(f'{outdir}/kmeans{K}'):
os.mkdir(f'{outdir}/kmeans{K}')
utils.save_pkl(kmeans_labels, f'{outdir}/kmeans{K}/labels.pkl')
np.savetxt(f'{outdir}/kmeans{K}/centers.txt', centers)
np.savetxt(f'{outdir}/kmeans{K}/centers_ind.txt', centers_ind, fmt='%d')
log('Generating volumes...')
vg.gen_volumes(f'{outdir}/kmeans{K}', centers)
# UMAP -- slow step
if zdim > 2 and not skip_umap:
log('Running UMAP...')
umap_emb = analysis.run_umap(z)
utils.save_pkl(umap_emb, f'{outdir}/umap.pkl')
# Make some plots
log('Generating plots...')
plt.figure(1)
g = sns.jointplot(x=pc[:,0], y=pc[:,1], alpha=.1, s=2)
g.set_axis_labels('PC1','PC2')
plt.tight_layout()
plt.savefig(f'{outdir}/z_pca.png')
plt.figure(2)
g = sns.jointplot(x=pc[:,0], y=pc[:,1], kind='hex')
g.set_axis_labels('PC1','PC2')
plt.tight_layout()
plt.savefig(f'{outdir}/z_pca_hexbin.png')
if zdim > 2 and not skip_umap:
plt.figure(3)
g = sns.jointplot(x=umap_emb[:,0], y=umap_emb[:,1], alpha=.1, s=2)
g.set_axis_labels('UMAP1','UMAP2')
plt.tight_layout()
plt.savefig(f'{outdir}/umap.png')
plt.figure(4)
g = sns.jointplot(x=umap_emb[:,0], y=umap_emb[:,1], kind='hex')
g.set_axis_labels('UMAP1','UMAP2')
plt.tight_layout()
plt.savefig(f'{outdir}/umap_hexbin.png')
analysis.scatter_annotate(pc[:,0], pc[:,1], centers_ind=centers_ind, annotate=True)
plt.xlabel('PC1')
plt.ylabel('PC2')
plt.savefig(f'{outdir}/kmeans{K}/z_pca.png')
g = analysis.scatter_annotate_hex(pc[:,0], pc[:,1], centers_ind=centers_ind, annotate=True)
g.set_axis_labels('PC1','PC2')
plt.tight_layout()
plt.savefig(f'{outdir}/kmeans{K}/z_pca_hex.png')
if zdim > 2 and not skip_umap:
analysis.scatter_annotate(umap_emb[:,0], umap_emb[:,1], centers_ind=centers_ind, annotate=True)
plt.xlabel('UMAP1')
plt.ylabel('UMAP2')
plt.savefig(f'{outdir}/kmeans{K}/umap.png')
g = analysis.scatter_annotate_hex(umap_emb[:,0], umap_emb[:,1], centers_ind=centers_ind, annotate=True)
g.set_axis_labels('UMAP1','UMAP2')
plt.tight_layout()
plt.savefig(f'{outdir}/kmeans{K}/umap_hex.png')
for i in range(num_pcs):
if zdim > 2 and not skip_umap:
analysis.scatter_color(umap_emb[:,0], umap_emb[:,1], pc[:,i], label=f'PC{i+1}')
plt.xlabel('UMAP1')
plt.ylabel('UMAP2')
plt.tight_layout()
plt.savefig(f'{outdir}/pc{i+1}/umap.png')