我正在一个拥有28个CPU和~190GB RAM的集群节点上训练sklearn.ensemble.RandomForestClassifier()
。单独训练此分类器运行速度非常快,使用机器上的所有核心并使用~93GB RAM:
x_train,x_test,y_train,y_test=sklearn.model_selection.train_test_split(x,y,test_size=.25,random_state=0)
clf=sklearn.ensemble.RandomForestClassifier(n_estimators=100,
random_state=0,
n_jobs=-1,
max_depth=10,
class_weight='balanced',
warm_start=False,
verbose=2)
clf.fit(x_train,y_train)
输出:
[Parallel(n_jobs=-1)]: Done 88 out of 100 | elapsed: 1.9min remaining: 15.2s
[Parallel(n_jobs=-1)]: Done 100 out of 100 | elapsed: 2.0min finished
CPU times: user 43min 10s, sys: 1min 33s, total: 44min 44s
Wall time: 2min 20s
然而,这种特殊模型似乎不是最佳的,具有约80%的正确性能。所以我想使用sklearn.model_selection.RandomizedSearchCV()
优化模型的超参数。所以我像这样设置搜索:
rfc = sklearn.ensemble.RandomForestClassifier()
rf_random = sklearn.model_selection.RandomizedSearchCV(estimator=rfc,
param_distributions=random_grid,
n_iter=100,
cv=3,
verbose=2,
random_state=0,
n_jobs=2,
pre_dispatch=1)
rf_random.fit(x, y)
但我无法找到有效使用硬件的n_jobs
和pre_dispatch
的设置。以下是一些示例运行和结果:
n_jobs pre_dispatch Result
===========================================================================
default default Utilizes all cores but Job killed - out of memory
-1 1 Job killed - out of memory
12 1 Job killed - out of memory
3 1 Job killed - out of memory
2 1 Job runs, but only utilizes 2 cores, takes >230min (wall clock) per model
在运行超参数搜索时,如何获得我在训练独立RandomForestClassifier
时看到的性能?独立版本如何并行化,以便不像网格搜索那样创建我的大型数据集的副本?
编辑:以下参数组合有效地使用所有核心来训练每个单独的RandomForestClassifier
,而无需并行化超参数搜索本身或炸毁RAM使用。
model = sklearn.ensemble.RandomForestClassifier(n_jobs=-1, verbose=1)
search = sklearn.model_selection.RandomizedSearchCV(estimator=model,
param_distributions=random_grid,
n_iter=10,
cv=3,
verbose=10,
random_state=0,
n_jobs=1,
pre_dispatch=1)
with joblib.parallel_backend('threading'):
search.fit(x, y)
如果单个分类器训练使所有核心饱和,那么通过并行化gridsearch也没有任何好处。为gridsearch设置n_jobs = 1,并为分类器保留n_jobs = -1。这应该避免内存不足的情况。