All Questions
322 questions
0
votes
0
answers
552
views
Theano AttributeError: 'TensorVariable' object has no attribute 'rank'
I'm currently reproducing this code
In the transform_rnn file, I have an issue.
class augmentaion(Layer):
def __init__(self,**kwargs):
self.supports_masking = True
self.uses_learning_phase = ...
- 95
2
votes
1
answer
2k
views
How to check if Keras backend is tensorflow or theano
I want to check in code what backend keras is using. Whether it is tensorflow or theano? How can I get this?
The code here uses K._BACKEND, however I'm getting an error saying that attribute not found....
- 7,297
0
votes
1
answer
1k
views
cannot import name 'theano_backend' from 'keras.backend'
I am trying to run the following:
from keras.backend import theano_backend
But I get this error:
Traceback (most recent call last):
File "<ipython-input-64-39e623866e51>", line 1, ...
0
votes
0
answers
37
views
Tensorflow isn't running multiple on CPUs, even though it is supposed to
I'm currently using an instance of Ubuntu 18.04.5 LTS using a google cloud service.
There's some code that I'm trying to use and for some reason, tensorflow won't use all of the instance's eight CPUs.
...
- 1
0
votes
1
answer
1k
views
NotImplementedError: Cannot convert a symbolic Tensor
I am trying to make a custom layer. My layer code is,
from keras import backend as K
from theano import tensor as T
from theano.tensor.signal.conv import conv2d
from keras.engine.topology import Layer
...
- 1
1
vote
0
answers
1k
views
tensorflow.python.util.tf_export.symbolalreadyexposederror: symbol name_scope_v1 is already exposed as ()
I am trying to set up theano backend for keras in my django server. And i got this error. I have no idea about this error please help me.
(func.__name__, getattr(func, api_names_attr))) # pylint: ...
- 115
0
votes
1
answer
1k
views
Keras isn't using Theano
1- I create a virtual env: mkvirtualenv kerasTH
2- I install keras using pip install keras
3- That's the output for pip list
Package Version
------------- -------
h5py 2.10.0
joblib ...
- 145
0
votes
0
answers
1k
views
What is the function of the InputSpec layer in Keras?
I came across layer 'InputSpec' in a Keras code which was used in the following form in the code -
from keras.layers import InputSpec
var = InputSpec(ndim=4)
The official documentation says it '...
- 386
2
votes
0
answers
76
views
Switching from Theano 1.0.0 to TensorFlow 2.2 (both using Keras) causes exponentially growing loss and stagnant accuracy
System information
Have I written custom code (as opposed to using a stock example script provided in TensorFlow): Yes
OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Linux Ubuntu 16.04
...
- 21
0
votes
0
answers
306
views
How can I change the backend of Keras dynamically?
Preliminary information: The answers given for the previously asked related questions (e.g., How to change the backend of Keras to Theano?, How to switch Backend with Keras (from TensorFlow to Theano))...
- 6,456
0
votes
1
answer
706
views
Assertion error when loading keras model with theano backend
I am having trouble loading a model built using keras with a theano backend. I am using Python 2, keras version 2.3.1, theano version 1.0.4. The model is built, trained, and saved as such:
import ...
- 180
3
votes
1
answer
2k
views
I dont understand what is wrong InvalidArgumentError: Conv2DCustomBackpropInputOp only supports NHWC
I am trying to execute this example notebook on modulation https://github.com/radioML/examples/blob/master/modulation_recognition/RML2016.10a_VTCNN2_example.ipynb
After executing this
from ...
- 41
0
votes
0
answers
56
views
How to view the values of a tensor and perform indexing on tensors when using Keras with Theano backend?
I'm trying to reproduce a network where points in the embedded space (of the autoencoder) need to be updated between each batch using Keras with a Theano backend. For the network to work, it needs to ...
- 92
0
votes
0
answers
4k
views
TypeError: predict() missing 1 required positional argument: 'X'
I'm trying to use scikit-learn in conjunction with Keras (Theano backend) in order to optimize some hyperparameters of my neural network, however I keep getting the same error (as stated in the title)....
0
votes
1
answer
508
views
from Keras.layers.core import Dense using theano as backend when creating a model
I have imported Dense from keras.layers.core ,When I create a new model
model = Sequential()
model.add(Dense(12, input_shape=(len(train_x[0]),), activation='relu'))
model.add(Dropout(0.5))
model.add(...
