由于某些原因，来自sklearn.ensemble的RandomForestClassifier.fit在我的本地机器上只使用了2.5GB的内存，但在我的服务器上使用了几乎7GB的内存，训练集完全相同。
不包括导入的代码大致如下：
y_train = data_train['train_column']x_train = data_train.drop('train_column', axis=1)# 内存消耗差异从这里开始clf = RandomForestClassifier(n_estimators=100, random_state=42)clf = clf.fit(x_train, y_train)preds = clf.predict(data_test)我的本地机器是配备16GB内存和4核CPU的MacBook Pro，我的服务器是DigitalOcean云上的Ubuntu服务器，同样配备8GB内存和4核CPU。
scikit-learn的版本是0.18，Python的版本是3.5.2
我甚至无法想象可能的原因，任何帮助都会非常有用。
更新
内存错误出现在fit方法内的以下代码中：
# 并行循环：我们使用线程后端，因为Cython代码# 用于拟合树木内部释放了Python的GIL，# 在这种情况下，线程总是比多处理更有效。trees = Parallel(n_jobs=self.n_jobs, verbose=self.verbose, backend="threading")( delayed(_parallel_build_trees)( t, self, X, y, sample_weight, i, len(trees), verbose=self.verbose, class_weight=self.class_weight) for i, t in enumerate(trees))更新2
关于我的系统的信息：
# 本地Darwin-16.1.0-x86_64-i386-64bitPython 3.5.2 (default, Oct 11 2016, 05:05:28)[GCC 4.2.1 Compatible Apple LLVM 8.0.0 (clang-800.0.38)]NumPy 1.11.2SciPy 0.18.1Scikit-Learn 0.18# 服务器Linux-3.13.0-57-generic-x86_64-with-Ubuntu-16.04-xenialPython 3.5.1 (default, Dec 18 2015, 00:00:00)[GCC 4.8.4]NumPy 1.11.2SciPy 0.18.1Scikit-Learn 0.18还有我的NumPy配置：
# 服务器>>> np.__config__.show()blas_opt_info: libraries = ['openblas', 'openblas'] define_macros = [('HAVE_CBLAS', None)] library_dirs = ['/usr/local/lib'] language = copenblas_info: libraries = ['openblas', 'openblas'] define_macros = [('HAVE_CBLAS', None)] library_dirs = ['/usr/local/lib'] language = clapack_opt_info: libraries = ['openblas', 'openblas'] define_macros = [('HAVE_CBLAS', None)] library_dirs = ['/usr/local/lib'] language = cblas_mkl_info: NOT AVAILABLEopenblas_lapack_info: libraries = ['openblas', 'openblas'] define_macros = [('HAVE_CBLAS', None)] library_dirs = ['/usr/local/lib'] language = c# 本地>>> np.__config__.show()blas_opt_info: extra_link_args = ['-Wl,-framework', '-Wl,Accelerate'] define_macros = [('NO_ATLAS_INFO', 3), ('HAVE_CBLAS', None)] extra_compile_args = ['-msse3', '-I/System/Library/Frameworks/vecLib.framework/Headers']blas_mkl_info: NOT AVAILABLEatlas_threads_info: NOT AVAILABLElapack_mkl_info: NOT AVAILABLEopenblas_lapack_info: NOT AVAILABLEatlas_info: NOT AVAILABLEatlas_3_10_blas_info: NOT AVAILABLElapack_opt_info: extra_link_args = ['-Wl,-framework', '-Wl,Accelerate'] define_macros = [('NO_ATLAS_INFO', 3), ('HAVE_CBLAS', None)] extra_compile_args = ['-msse3']openblas_info: NOT AVAILABLEatlas_3_10_blas_threads_info: NOT AVAILABLEatlas_3_10_threads_info: NOT AVAILABLEatlas_3_10_info: NOT AVAILABLEatlas_blas_threads_info: NOT AVAILABLEatlas_blas_info: NOT AVAILABLEclf对象在两台机器上的表示是相同的：
RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini', max_depth=None, max_features='auto', max_leaf_nodes=None, min_impurity_split=1e-07, min_samples_leaf=1, min_samples_split=2, min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=1, oob_score=False, random_state=42, verbose=0, warm_start=False)回答：
好吧，在我将内核从3.13.0-57更新到4.4.0-28后，问题神奇地消失了。现在它消耗的内存甚至比我的本地Mac笔记本电脑还要少。
