# local
import utils
# convert test statistic to a p-value for a given point
pVal_func = lambda TS, dec: -np.log10(0.5 * (chi2(len(llh.params)).sf(TS)
+ chi2(len(llh.params)).cdf(-TS)))
label = dict(TS=r"$\mathcal{TS}$",
nsources=r"$n_S$",
gamma=r"$\gamma$",
)
if __name__=="__main__":
plt = utils.plotting(backend="pdf")
llh, mc = utils.startup(Nsrc=10)
print(llh)
# iterator of all-sky scan with follow up scans of most interesting points
for i, (scan, hotspot) in enumerate(llh.all_sky_scan(
nside=2**6, follow_up_factor=1,
pVal=pVal_func,
decRange=np.radians([-90., 90.]))):
if i > 0:
# break after first follow up
break
logging.getLogger("skylab.psLLH.PointSourceLLH").setLevel(logging.INFO)
# convert test statistic to a p-value for a given point
pVal_func = lambda TS, dec: -np.log10(0.5 * (chi2(len(llh.params)).sf(TS)
+ chi2(len(llh.params)).cdf(-TS)))
label = dict(TS=r"$\mathcal{TS}$",
nsources=r"$n_S$",
gamma=r"$\gamma$",
)
if __name__=="__main__":
plt = utils.plotting(backend="pdf")
llh, mc = utils.startup(Nsrc=10)
print(llh)
# iterator of all-sky scan with follow up scans of most interesting points
for i, (scan, hotspot) in enumerate(llh.all_sky_scan(
nside=2**6, follow_up_factor=1,
pVal=pVal_func,
hemispheres=dict(Full=np.radians([-90., 90.])))):
if i > 0:
# break after first follow up
break
from scipy.interpolate import interp1d
from sklearn.linear_model import LinearRegression
from sklearn.preprocessing import PolynomialFeatures
from sklearn.metrics import mean_squared_error, r2_score
from sklearn.metrics import accuracy_score
samplerate, samples = wavfile.read('canciones/hakuna_matata.wav')
samples = samples[5000000:5000100]
newsamples = samples.copy()
damage.noiseadd(newsamples, 0.7, 0.3)
matches = recognize.cheat(samples,
newsamples,
false_positives=0.04,
false_negatives=0.1)
matchesSD = recognize.cheat(samples,
samples,
false_positives=0.04,
false_negatives=0.1)
x, y = utils.tovalidxy(newsamples, matches)
xSD, ySD = utils.tovalidxy(samples, matchesSD)
x = np.array(x).reshape((-1, 1))
y = np.array(y)
xSD = np.array(xSD).reshape((-1, 1))
ySD = np.array(ySD)
xP, yP, xSD, ySD = utils.partir(x, y, xSD, ySD, 10)
polynomial_features = PolynomialFeatures(degree=10)
y_poly_pred = utils.polinomialR(xP, yP, ySD, polynomial_features)
utils.plotting(xP, yP, samples, y_poly_pred)
def map_states(a):
set_a = np.array(list(set(a)))
for ii in range(len(set_a)):
a = np.where(a==set_a[ii],ii+1,a)
return a
if __name__ == '__main__':
for target in targets:
out_dir = os.path.join(ouput_prefix, target)
if not os.path.exists(out_dir):
os.makedirs(out_dir)
out_edf_data = load_edf_data(input_prefix, target)
counter_data = 1
Theta = None
U = None
for data_all, file_name, seizure_start_time_offsets, seizure_lengths in out_edf_data:
# plot_eeg(data_all, seizure_start_time_offsets, seizure_lengths)
counter_data, weight_matrices_features, weight_matrices,intervals_seizures, Theta, U = \
W_generator(data_all, file_name, seizure_start_time_offsets, seizure_lengths, target, out_dir, out_edf_data, counter_data, flag_normalized, Theta_in=Theta, U_in=U)
estimated_states = None
for appr in range(num_approaches):
# estimated_states = state_estimation(weight_matrices_features[appr,:,:], file_name, state_estimation_mode)
# estimated_states = map_states(estimated_states)
plotting(out_dir + '/' + name_weight_matrices[appr], file_name, weight_matrices[appr], intervals_seizures = intervals_seizures, estimated_states = estimated_states)
def main_():
N = 15
#Sector Basic Materials
syms = BM
sect = sector(syms)
syms = random.sample(syms, N)
plotting((pre_main_(syms)), sect)
#'''
#Sector Communication Services
syms = CS
sect = sector(syms)
syms = random.sample(syms, N)
plotting((pre_main_(syms)), sect)
#Sector Consumer Defensive
syms = CD
sect = sector(syms)
syms = random.sample(syms, N)
plotting((pre_main_(syms)), sect)
#Sector Energy
syms = Eneg
sect = sector(syms)
syms = random.sample(syms, N)
plotting((pre_main_(syms)), sect)
#Sector Real Estate
syms = RE
sect = sector(syms)
plotting((pre_main_(syms)), sect)
#Sector Industrials
syms = Ind
sect = sector(syms)
syms = random.sample(syms, N)
plotting((pre_main_(syms)), sect)
######################
#Sector Consumer Cyclical
syms = CC
sect = sector(syms)
syms = random.sample(syms, N)
plotting((pre_main_(syms)), sect)
#Sector Financial Services
syms = FS
sect = sector(syms)
syms = random.sample(syms, N)
plotting((pre_main_(syms)), sect)
#Sector Health Care
syms = HC
sect = sector(syms)
syms = random.sample(syms, N)
plotting((pre_main_(syms)), sect)
#Sector Technology
syms = Techn
sect = sector(syms)
syms = random.sample(syms, N)
plotting((pre_main_(syms)), sect)
df_ts = utils.rm_early_zeros(df_ts)
#if max_display_length > 0:
# df_ts = df_ts[-max_display_length:]
results, model = utils.process_geounit(df_ts, window_length, exp_or_lin)
print()
utils.print_results(selection, results.iloc[0, 0:8], normalise_by, pop_csv,
results.iloc[0,8], results.iloc[0,9], case, lang)
if save_not_show in [0, 1]:
country = selection + ' '
if case=='confirmed':
country += 'Fallzahl'
else:
country += 'Todesfälle'
utils.plotting(df_ts, model, save_not_show, country, results.iloc[0,8],
results.iloc[0,9], lang, panels)
else: # analysis of all counties and complete BW
for case in cases_list:
if max_display_length > 0:
figures[case] = figures[case].iloc[-max_display_length:,:]
results_list = list()
#for selection in allowed_values[:3]:
for selection in allowed_values[:-1]:
print(selection)
df_ts = figures[case][selection]
df_ts = utils.rm_early_zeros(df_ts)
results, model = utils.process_geounit(df_ts, window_length, exp_or_lin)
results = results.assign(selection=selection)
if __name__ == '__main__':
pop_csv = 'world'
df = utils.open_csvs() # filename is used to create name of image file if saving plot
results_list = list()
for country in countries:
print(country)
df_ts = utils.data_preparation(df, country, cases)
if max_display_length > 0:
df_ts = df_ts[-max_display_length:]
results, model = utils.process_geounit(df_ts, window_length, exp_or_lin)
results = results.assign(country=country)
results = results.set_index('country')
results_list.append(results)
if save_plots == 1:
utils.plotting(df_ts, model, 1, country, results.iloc[0,8], results.iloc[0,9], lang, panels)
utils.print_header(normalise_by, pop_csv)
df_results = pd.concat(results_list)
df_results = df_results.sort_values(3, ascending=False)
for country in df_results.index:
if window_length > 0:
utils.print_results(country, df_results.loc[country,:].iloc[0:8], normalise_by, pop_csv,
window_length, df_results.loc[country,:].iloc[9], cases, lang)
else:
utils.print_results(country, df_results.loc[country,:].iloc[0:8], normalise_by, pop_csv,
df_results.loc[country,:].iloc[8], df_results.loc[country,:].iloc[9], cases, lang)