The pilots's discuss from rpi-wcl

pilots's Issues

machine learning engine execution error

This error occurs on Ubuntu 14.04.5 with Python version 2.7.10. When attempting to execute:

python engineIO.py definitions/bayes_config.json

from pilots/util/learningmodel/engine on the learn_dev branch as an example of using the machine learning engine, the following error output message occurs:

Traceback (most recent call last):
  File "engineIO.py", line 100, in <module>
    env = train_model(json_file)
  File "engineIO.py", line 14, in train_model
    return _load(json_file_name)
  File "engineIO.py", line 87, in _load
    read_model(env, json_dict[K.M])
  File "engineIO.py", line 57, in read_model
    env.feature = env.transform(baseenv.BaseEnv.data_transformer, schema=env.get_schema_info(), constant=env.get_constant(), definition=json_dict[K.A_FEATURES])
  File "/home/glownd/Documents/PILOTS_research/git_repos/pilots/pilots/util/learningmodel/engine/scbase/baseenv.py", line 125, in transform
    return transformer(self.data, **parameters)
  File "/home/glownd/Documents/PILOTS_research/git_repos/pilots/pilots/util/learningmodel/engine/scbase/baseenv.py", line 196, in data_transformer
    X = np.apply_along_axis(def_transformer, 1, npmatrix)
  File "/usr/local/lib/python2.7/dist-packages/numpy/lib/shape_base.py", line 91, in apply_along_axis
    res = func1d(arr[tuple(i.tolist())], *args, **kwargs)
  File "<string>", line 1, in <lambda>
  File "/usr/local/lib/python2.7/dist-packages/numpy/matrixlib/defmatrix.py", line 318, in __getitem__
    out = N.ndarray.__getitem__(self, index)
IndexError: index 3 is out of bounds for axis 0 with size 1

Remove usage of java -cp everywhere

Instead of using java -cp in the example files to compile the generated java files, it's easier to set up the java class path directly in setenv file.

Recommend Projects