import scimpute
import Orange.data
from Orange.data import Table
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.colors as colors
# how to load files into numpy
# file_inflated = np.genfromtxt('biological_inflated.csv', delimiter=',')
# Load
filename = "../../data/ccp_normCounts_mESCquartz.counts.cycle_genes.csv"
dataset = Orange.data.Table(filename)
# Izracunaj vse potrebno
dat, mas, zero = scimpute.zero_inflate(dataset.X)
sc = scimpute.ScImpute(dataset.X)
res = sc.scvis(dat)
cor, data = sc.compare_embedded(res)
print(cor)
razlika1 = []
razlika2 = []
for x in range(len(data[0])):
if (data[0][x] > 0.7 or data[0][x] < -0.7):
razlika1.append(data[1][x] - data[0][x])
if (data[0][x] < 0.3 and data[0][x] > -0.3):
razlika2.append(data[1][x] - data[0][x])
# Plotaj vse potrebno
# Primerjava bioloskih podatkov z imputiranimi vrednostmi
import sys
sys.path.append("../..")
import scimpute
import Orange.data
from Orange.data import Table
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.colors as colors
# Izracunaj vse potrebno z uporabo modula
data_org = scimpute.generate()
dat, mas, zero = scimpute.zero_inflate(data_org)
sc = scimpute.ScImpute(dat)
res = sc.WMean_chisquared()
cor, data = sc.compare(data_org, mas)
print(cor)
# Plotaj vse potrebno
# Primerjava bioloskih podatkov z imputiranimi vrednostmi
fig, (ax0, ax1, ax2) = plt.subplots(3, 1)
c = ax0.pcolormesh(data_org, norm=colors.LogNorm(vmin=np.amin(data_org)+1, vmax=np.amax(data_org)), cmap=plt.get_cmap("binary"))
fig.colorbar(c, ax=ax0)
ax0.set_title('Sintetični podatki')
c = ax1.pcolormesh(res,norm=colors.LogNorm(vmin=np.amin(res)+1, vmax=np.amax(res)), cmap=plt.get_cmap("binary"))
ax1.set_title('Imputirani sintetični podatki')
fig.colorbar(c, ax=ax1)
c = ax2.pcolormesh(data_org-res, norm=colors.LogNorm(vmin=np.amin(data_org)+1, vmax=np.amax(data_org)), cmap=plt.get_cmap("binary"))
ax2.set_title('Razlika med sintetičnimi in imputiranimi podatki')