TensorFlow: Combining Categorical and Continuous Variables
Illia Polosukhin
2017

Hi, Could you please help with updating it for TensorFlow 1.0. I was trying to run my script (that work on TF 0.10). But, with TF 1.0, it fails with the following error:
 File “online_tensor1_Hotel.py”, line 289, in <module>
 classifier.fit(input_fn=pandas_input_fn(X_train.loc[0:train_base,],y_train.loc[0:train_base]), steps=2500,monitors=[validation_monitor])
 File “/usr/lib/python2.7/dist-packages/tensorflow/python/util/deprecation.py”, line 280, in new_func
 return func(*args, **kwargs)
 File “/usr/lib/python2.7/dist-packages/tensorflow/contrib/learn/python/learn/estimators/estimator.py”, line 426, in fit
 loss = self._train_model(input_fn=input_fn, hooks=hooks)
 File “/usr/lib/python2.7/dist-packages/tensorflow/contrib/learn/python/learn/estimators/estimator.py”, line 932, in _train_model
 features, labels = input_fn()
 File “online_tensor1_Hotel.py”, line 168, in input_fn
 queue = learn.DataFrame.queues.feeding_functions.enqueue_data(
AttributeError: ‘module’ object has no attribute ‘DataFrame’

Would you have any suggestions?