stop = lambda ind: ind.fitness.values[0] >= 0.95
sl = SymbolLearning(loop="MultiMutateLoop", pset=pset0, gen=20, pop=1000, hall=1, batch_size=40, re_hall=3,
n_jobs=12, mate_prob=0.9, max_value=h_bgp, initial_min=2, initial_max=h_bgp,
mutate_prob=0.8, tq=False, dim_type="coef", stop_condition=stop,
re_Tree=0, store=False, random_state=4, verbose=True,
# stats=None,
stats={"fitness_dim_max": ["max"], "dim_is_target": ["sum"], "h_bgp": ["mean"]},
add_coef=True, inter_add=True, out_add=True, cal_dim=True, vector_add=True,
personal_map=False)
tt.t
sl.fit()
tt.t
tt.p
score = sl.score(x, y, "r2")
print(sl.expr)
y_pre = sl.predict(x)
break
# just for shown
# y_pre = si_transformer.scale_y * y_pre
# ssc = Dim.inverse_convert(y_dim, target_units=eV)[0]
# y_pre = y_pre * ssc
#
# p = BasePlot(font=None)
# p.scatter(Y, y_pre, strx='Experimental $E_{gap}$', stry='Calculated $E_{gap}$')
# import matplotlib.pyplot as plt
#
# plt.show()
from sklearn.linear_model import LinearRegression
tq=True,
dim_type="coef",
stop_condition=stop,
re_Tree=0,
random_state=0,
verbose=True,
add_coef=True,
inter_add=True,
cal_dim=False,
inner_add=True,
personal_map=False)
# sl.fit()
print(sl.expr)
y_pre = sl.predict(X)
score_all = sl.score(X, y, "r2")
p = BasePlot(font=None)
p.scatter(y,
y_pre,
strx='Experimental $E_{gap}$',
stry='Calculated $E_{gap}$')
import matplotlib.pyplot as plt
plt.show()
y_pre = sl.predict(X_test)
score_test = sl.score(X_test, y_test, "r2")
p = BasePlot(font=None)
],
cal_dim=True,
inner_add=True,
out_add=False,
n_jobs=10,
)
#
# 方式选择4，系数加在公式内层,不考虑量纲计算,
# sl = SymbolLearning(loop="MultiMutateLoop", pop=200, gen=10, random_state=1,pset=pset0,
# classification=True, scoring=[metrics.accuracy_score, ], score_pen=[1, ],
# cal_dim=False,
# inner_add=False, out_add=False,
# n_jobs = 2,)
# #
# # 方式选择5，,公式项全部展开，每一项前面加系数（非常复杂),不考虑量纲计算,
# pset0.y_dim = None
# sl = SymbolLearning(loop="MultiMutateLoop", pop=1000, gen=20, random_state=1,pset=pset0,
# classification=True, scoring=[metrics.accuracy_score, ], score_pen=[1, ],
# cal_dim=False, flat_add=True, out_add=False,
# n_jobs = 10,)
#
# # 方式选择6，,公式项全部展开，每一项前面加系数（非常复杂),不考虑量纲计算, 公式复杂度不限制
# pset0.y_dim = None
# sl = SymbolLearning(loop="MultiMutateLoop", pop=1000, gen=10, random_state=1, max_value=7,pset=pset0,
# classification=True, scoring=[metrics.accuracy_score, ], score_pen=[1, ],
# cal_dim=False, flat_add=True, out_add=False,
# n_jobs = 10,)
sl.fit(x, y)
pre_y = sl.predict(x)
print(sl.expr)