def sample_export_R (self,anzgrids,fname,vx,v1,v2=None,v3=None): #print "sample_export_R: anzgrids=%d, vx=%s,v1=%s,v2=%s,v3=%s" #% (anzgrids,vx,v1,v2,v3) g2 = None g3 = None gx = self.d_grids[vx] # Pflicht if gx is None: return None, None g1 = self.d_grids[v1] # Pflicht if g1 is None: return None, None n = 2000 anz = int(anzgrids) if anz == 2: res, resy = sample(g1,g4=gx,n=n,filename=fname,x1=v1,x4=vx) elif anz == 3: if v2 is not None: g2 = self.d_grids[v2] if g2 is not None: return None, None res, resy = sample(g1,g4=gx,g2=g2,n=n, filename=fname,x1=v1,x2=v2,x4=vx) elif anz == 4: if v2 is not None: g2 = self.d_grids[v2] if g2 is None: return None, None if v3 is not None: g3 = self.d_grids[v3] if g3 is None: return None, None res, resy = sample(g1,g4=gx,g2=g2,g3=g3,n=n, filename=fname,x1=v1,x2=v2,x3=v3,x4=vx) return res, resy
def test():
#~ #define the grids
bd = grid.grid()
struktur = grid.grid()
nahr = grid.grid()
# load the grids
bd.read_hdf('../../hdf/schreiadler_all.h5', 'bd_all')
struktur.read_hdf('../../hdf/schreiadler_all.h5', 'struktur')
nahr.read_hdf('../../hdf/schreiadler_all.h5', 'nahr_schreiadler')
#~ # draw a sample of a size of 2000
X, Y = grid.sample(g1=bd,
g2=struktur,
g4=nahr,
n=2000,
filename='s_nahr.csv')
X = np.array(X)
Y = np.array(Y)
print X.shape, Y.shape
# define the model
t0 = time.time()
model = svm('bd', 'struktur')
model.def_model()
# train the model
model.train(X, Y)
# test model
bias, rsme, r2 = model.test(X, Y)
print 'Bias=', bias, 'RSME=', rsme, 'R^2=', r2
# write the model
model.write_model('bd_struk.svm')
print 'after read:', model.get_names()
# gen the output
out = model.calc(bd, struktur)
print 'time=', time.time() - t0
out.show()
def test():
#~ #define the grids
bd=grid.grid()
struktur=grid.grid()
nahr=grid.grid()
# load the grids
bd.read_hdf('../../hdf/schreiadler_all.h5','bd_all')
struktur.read_hdf('../../hdf/schreiadler_all.h5','struktur')
nahr.read_hdf('../../hdf/schreiadler_all.h5','nahr_schreiadler')
#~ # draw a sample of a size of 2000
X,Y=grid.sample(g1=bd,g2=struktur,g4=nahr,n=2000,filename='s_nahr.csv')
X=np.array(X)
Y=np.array(Y)
print X.shape, Y.shape
# define the model
t0=time.time()
model=svm('bd','struktur')
model.def_model()
# train the model
model.train(X,Y)
# test model
bias,rsme,r2=model.test(X,Y)
print 'Bias=', bias, 'RSME=', rsme, 'R^2=', r2
# write the model
model.write_model('bd_struk.svm')
print 'after read:', model.get_names()
# gen the output
out=model.calc(bd,struktur)
print 'time=',time.time()-t0
out.show()
def main():
# Create a random dataset
rng = np.random.RandomState(1)
x = np.sort(5 * rng.rand(64 * 4, 1), axis=0)
y = np.sin(x).ravel()
y[::64] += 3 * (0.5 - rng.rand(4))
data = np.array([x, y]).transpose()
size = int(np.log2(data.shape[0])) ** 2
data, w = grid.sample(data, size)
x_1, y_1 = data[:, 0].reshape((size, 1)), data[:, 1].reshape((size, 1))
i = range(size)
i = np.random.choice(i, size)
x_2, y_2 = x[i], y[i]
clf_1 = DecisionTreeRegressor(max_depth=D)
clf_1.fit(x_1, y_1, sample_weight=w)
clf_2 = DecisionTreeRegressor(max_depth=D)
clf_2.fit(x_2, y_2)
# Predict
X_test = np.arange(0.0, 5.0, 0.01)[:, np.newaxis]
y_1 = clf_1.predict(X_test)
y_2 = clf_2.predict(X_test)
# Plot the results
plt.figure()
plt.scatter(x, y, c="k", label="data")
plt.plot(X_test, y_1, c="g", label="coreset", linewidth=2)
plt.plot(X_test, y_2, c="r", label="random", linewidth=2)
plt.xlabel("data")
plt.ylabel("target")
plt.title("Decision Tree Regression")
plt.legend()
plt.show()
