Question

我有一个名为X的DataFrame和一组名为Y的目标值。

对于我的大多数模特，我都会做这样的事情（只是一个例子）：

from sklearn.linear_model import LassoCV
clf = LassoCV()
score = cross_val_score(estimator = clf, X = X, y = Y, cv = KFold(n_splits = 3, random_state = 100), n_jobs = -1, \
                        scoring = "neg_mean_squared_error")
np.mean([np.sqrt(-x) for x in score])

我试图以类似的方式使用TPOT，如下所示：

from tpot import TPOTRegressor
tpot = TPOTRegressor(generations=20, population_size=100, verbosity=2)

score = cross_val_score(estimator = tpot, X = X, y = Y, cv = KFold(n_splits = 3, random_state = 100), n_jobs = -1, \
                        scoring = "neg_mean_squared_error")
np.mean([np.sqrt(-x) for x in score])

TPOT启动但后来给我一个酸洗错误如下：

PicklingError: Can't pickle <type 'instancemethod'>: it's not found as __builtin__.instancemethod

知道为什么会这样发生/怎么让TPOT好好玩？

谢谢！

Answer 1

如果您使用的是Python 2，尝试：

import dill

所以lambda函数可以被腌制....为我工作......

在Python 3中，您可能需要：

import dill as pickle

Answer 2

尝试使用： tpot.fitted_pipeline _

from tpot import TPOTRegressor
tpot = TPOTRegressor(generations=20, population_size=100, verbosity=2)

score = cross_val_score(estimator = tpot.fitted_pipeline_, X = X, y = Y, cv = KFold(n_splits = 3, random_state = 100), n_jobs = -1, \
                        scoring = "neg_mean_squared_error")
np.mean([np.sqrt(-x) for x in score])

TPOT：使用TPOTRegressor时出现酸洗错误

2 个答案: