TypeError: 'generator' object is not subscriptable

[nguyenhoamy1602]

When running Iris Example and MNIST Example. The following error is seen:

Traceback (most recent call last):
  File "1.py", line 31, in <module>
    estim.fit( X_train, y_train )
  File "/Users/nguyen.hoamy/PycharmProjects/ML/hyperopt-sklearn/hpsklearn/estimator.py", line 747, in fit
    fit_iter.send(increment)
  File "/Users/nguyen.hoamy/PycharmProjects/ML/hyperopt-sklearn/hpsklearn/estimator.py", line 657, in fit_iter
    return_argmin=False, # -- in case no success so far
  File "/Users/nguyen.hoamy/PycharmProjects/ML/venv/lib/python3.6/site-packages/hyperopt/fmin.py", line 307, in fmin
    return_argmin=return_argmin,
  File "/Users/nguyen.hoamy/PycharmProjects/ML/venv/lib/python3.6/site-packages/hyperopt/base.py", line 635, in fmin
    return_argmin=return_argmin)
  File "/Users/nguyen.hoamy/PycharmProjects/ML/venv/lib/python3.6/site-packages/hyperopt/fmin.py", line 314, in fmin
    pass_expr_memo_ctrl=pass_expr_memo_ctrl)
  File "/Users/nguyen.hoamy/PycharmProjects/ML/venv/lib/python3.6/site-packages/hyperopt/base.py", line 786, in __init__
    pyll.toposort(self.expr)
  File "/Users/nguyen.hoamy/PycharmProjects/ML/venv/lib/python3.6/site-packages/hyperopt/pyll/base.py", line 715, in toposort
    assert order[-1] == expr
TypeError: 'generator' object is not subscriptable

[bjkomer]

This is due to an older version of hyperopt not working with the latest networkx.

You can either upgrade to the latest master of hyperopt using pip install --upgrade git+git://github.com/hyperopt/hyperopt.git
or downgrade the networkx dependency using pip install networkx==1.11

Duplicate of #90 and hyperopt issue 325