Parallelizing Evaluations During Search via MongoDB

Hyperopt is designed to support different kinds of trial databases. The default trial database (Trials) is based on Python lists and dictionaries, and does not support the asynchronous updates required for the parallel evaluation of trials. However, hyperopt also comes with a MongoTrials trials implementation that does support asynchronous updates, and the parallel evaluation of trials.

Parallel fmin() with MongoTrials

import hyperopt
from hyperopt import fmin, tpe, hp

from hyperopt import Trials
from hyperopt.mongoexp import MongoTrials

trials = Trials()

trials2 = MongoTrials('mongo://localhost:1234/my_db')

def fn(x, y):
     return {
         'loss': x ** 2 + y ** 2,
         'status': hyperopt.STATUS_OK,
     }

fmin(fn,
    space=[hp.uniform('a', -3, 3), hp.normal('b', 1, 1)],
    algo=tpe.suggest,
    max_evals=100,
    #trials=trials,  # -- runs locally, synchronously
    trials=trials2,  # -- runs async
    )

print trials