gpu - Multiprocessing for Python parallelization error - "function' object is not iterable" -

we have nvidia tesla k80 gpu accelerator computing in our data center following characteristics: intel(r) xeon(r) cpu e5-2670 v3 @2.30ghz, 48 cpu processors, 128gb ram, 12 cpu coresrunning under linux 64-bit.

i running following code gridsearchcv after vertically appending different sets of dataframes single series of data randomforestregressor model. example, 2 sample datasets considering found in this link

from joblib import parallel, delayed import multiprocessing import sys import imp import glob import os import pandas pd import math sklearn.feature_extraction.text import countvectorizer  sklearn.feature_extraction.text import tfidfvectorizer sklearn.linear_model import logisticregression import matplotlib sklearn.model_selection import cross_val_score import matplotlib.pyplot plt import numpy np sklearn.ensemble import randomforestregressor, gradientboostingregressor sklearn.naive_bayes import gaussiannb sklearn.model_selection import gridsearchcv sklearn.naive_bayes import multinomialnb sklearn.linear_model import lassocv sklearn.metrics import r2_score, mean_squared_error, make_scorer sklearn.model_selection import train_test_split math import sqrt sklearn.cross_validation import train_test_split  df = pd.concat(map(pd.read_csv, glob.glob(os.path.join('', "cubic*.csv"))), ignore_index=true) #df = pd.read_csv('cubic31.csv')  in range(1,3):     df['x_t'+str(i)] = df['x'].shift(i)  print(df)  df.dropna(inplace=true)  x = (pd.dataframe({ 'x_%d'%i : df['x'].shift(i) in range(3)}).apply(np.nan_to_num, axis=0).values)  x = df.drop('y', axis=1) y = df['y']  x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.40)  x_train = x_train.drop('time', axis=1) x_test = x_test.drop('time', axis=1)   print(x.shape) print(df['y'].shape)  print() print("size of x_train:",(len(x_train))) print("size of y_train:",(len(x_train))) print("size of x_test:",(len(x_test))) print("size of y_test:",(len(y_test)))  print()  def gridsearchcvparallel():     #fit models grid search cv=5 (not low), use best model     parameters = {'n_estimators': [10,30,100,500,1000]}     clf_rf = randomforestregressor(random_state=1)     clf = gridsearchcv(clf_rf, parameters, cv=5, scoring='neg_mean_squared_error')     model = clf.fit(x_train, y_train)     model.cv_results_['params'][model.best_index_]     math.sqrt(model.best_score_*-1)     model.grid_scores_      #####     print()     print(model.grid_scores_)     print("the best score: ",model.best_score_)      print("rmse:",math.sqrt(model.best_score_*-1))      #reg = randomforestregressor(criterion='mse')     clf_rf.fit(x_train,y_train)     modelprediction = clf_rf.predict(x_test)     print(modelprediction)      print("number of predictions:",len(modelprediction))      meansquarederror=mean_squared_error(y_test, modelprediction)     print("mean square error (mse):", meansquarederror)     rootmeansquarederror = sqrt(meansquarederror)     print("root-mean-square error (rmse):", rootmeansquarederror)       ####### add trendline     fig, ax = plt.subplots()     #df.plot(x='time', y='y', ax=ax)     ax.plot(df['time'].values, df['y'].values)       fig, ax = plt.subplots()     index_values=range(0,len(y_test))      y_test.sort_index(inplace=true)     x_test.sort_index(inplace=true)      modelpred_test = clf_rf.predict(x_test)     ax.plot(pd.series(index_values), y_test.values)       plotinone=pd.dataframe(pd.concat([pd.series(modelpred_test), pd.series(y_test.values)], axis=1))      plt.figure(); plotinone.plot(); plt.legend(loc='best') numberofcores = multiprocessing.cpu_count()  gridresults = parallel(n_jobs=numberofcores)(delayed(gridsearchcvparallel))  print(gridresults) 

when run program huge dataset (around 2 million rows), taking more 4 days gridsearchcv. after bit of search, found out python threads can utilize more 1 cpu using either concurrent.futures or multiprocessing. shown in code, tried make use of multiplrocessing getting error typeerror: 'function' object not iterable. seems function should take single parameter input , pass in iterable argument. how can fix issue utilize more 1 cpu , task faster in short period of time?

thank in advance.

do not attempt parallelize on own. not use joblib.parallel. reinventing wheel, anyway, since gridsearchcv already-parellized. pass n_jobs parameter, defaults 1, i.e. defaults using single job. take advantage of multi-core architecture, pass n_jobs = number_of_cores, number_of_cores number of cores want use.

and if check source code, you'll see wraps call joblib.parallel, n_jobs=-1 should work "all cores".