运行tensorflow例子遇到的问题和处理

========== RESTART: D:/pythonwork/tensorflow/example/5/nonlinear.py ==========
Traceback (most recent call last):
  File "D:/pythonwork/tensorflow/example/5/nonlinear.py", line 6, in <module>
    from sklearn import datasets, cross_validation, metrics
ImportError: cannot import name 'cross_validation'

#自0.18.1版本后sklearn移除了cross validation，因此下面改为model selection

>>>
========== RESTART: D:/pythonwork/tensorflow/example/5/nonlinear.py ==========
Traceback (most recent call last):
  File "D:/pythonwork/tensorflow/example/5/nonlinear.py", line 13, in <module>
    from keras.models import Sequential
ModuleNotFoundError: No module named 'keras'
>>>
========== RESTART: D:/pythonwork/tensorflow/example/5/nonlinear.py ==========
Using TensorFlow backend.
Traceback (most recent call last):
  File "D:/pythonwork/tensorflow/example/5/nonlinear.py", line 37, in <module>
    X_train, X_test, y_train, y_test = cross_validation.train_test_split(X, y,
NameError: name 'cross_validation' is not defined
>>>
========== RESTART: D:/pythonwork/tensorflow/example/5/nonlinear.py ==========
Using TensorFlow backend.

Warning (from warnings module):
  File "D:\Python36\lib\site-packages\sklearn\preprocessing\data.py", line 625
    return self.partial_fit(X, y)
DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.

Warning (from warnings module):
  File "D:\Python36\lib\site-packages\sklearn\base.py", line 462
    return self.fit(X, **fit_params).transform(X)
DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.

Warning (from warnings module):
  File "D:/pythonwork/tensorflow/example/5/nonlinear.py", line 52
    model.add(Dense(10, input_dim=7, init='normal', activation='relu'))
UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(10, input_dim=7, activation="relu", kernel_initializer="normal")`

Warning (from warnings module):
  File "D:/pythonwork/tensorflow/example/5/nonlinear.py", line 53
    model.add(Dense(5, init='normal', activation='relu'))
UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(5, activation="relu", kernel_initializer="normal")`

Warning (from warnings module):
  File "D:/pythonwork/tensorflow/example/5/nonlinear.py", line 54
    model.add(Dense(1, init='normal'))
UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(1, kernel_initializer="normal")`

Warning (from warnings module):
  File "D:/pythonwork/tensorflow/example/5/nonlinear.py", line 60
    model.fit(X_train, y_train, nb_epoch=1000, validation_split=0.33, shuffle=True,verbose=2 )
UserWarning: The `nb_epoch` argument in `fit` has been renamed `epochs`.PS

转载于:https://www.cnblogs.com/qianheng/p/10036281.html