Error in run with tensorflow

[AIAdventures]

Hi, was trying to test your code.

i was getting errors though.

andrewcz@andrewcz-PORTEGE-Z30t-B ~/Eigenvalue-Decay-Regularizer-for-Keras $ python example.py
Using TensorFlow backend.
60000 train samples
10000 test samples
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
Traceback (most recent call last):
File "example.py", line 64, in
metrics=['accuracy'])
File "/home/andrewcz/miniconda3/lib/python3.5/site-packages/keras/models.py", line 553, in compile
**kwargs)
File "/home/andrewcz/miniconda3/lib/python3.5/site-packages/keras/engine/training.py", line 641, in compile
total_loss = r(total_loss)
File "/home/andrewcz/Eigenvalue-Decay-Regularizer-for-Keras/EigenvalueDecay.py", line 57, in call
domin_eigenvect = K.dot(WW, o)
File "/home/andrewcz/miniconda3/lib/python3.5/site-packages/keras/backend/tensorflow_backend.py", line 340, in dot
if ndim(x) is not None and (ndim(x) > 2 or ndim(y) > 2):
File "/home/andrewcz/miniconda3/lib/python3.5/site-packages/keras/backend/tensorflow_backend.py", line 220, in ndim
dims = x.get_shape()._dims
AttributeError: 'numpy.ndarray' object has no attribute 'get_shape'

help please.
Many thanks,
Andrew

[oswaldoludwig]

Hi Andrew, This code is only for Theano backend, a version for both Theano and TensorFlow backends was available in Keras, but was removed in the new version of Keras: keras-team/keras@1c630c3 You can try to run my code using Theano backend, but I'm not sure if it will work with the current version of Keras, because I didn't upgrade my version of Keras yet. Best regards, Oswaldo 2017-04-26 6:25 GMT+02:00 AIAdventures <[email protected]>:

…

Hi, was trying to test your code. i was getting errors though. ***@***.*** ~/Eigenvalue-Decay-Regularizer-for-Keras $ python example.py Using TensorFlow backend. 60000 train samples 10000 test samples W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations. Traceback (most recent call last): File "example.py", line 64, in metrics=['accuracy']) File "/home/andrewcz/miniconda3/lib/python3.5/site-packages/keras/models.py", line 553, in compile **kwargs) File "/home/andrewcz/miniconda3/lib/python3.5/site-packages/keras/engine/training.py", line 641, in compile total_loss = r(total_loss) File "/home/andrewcz/Eigenvalue-Decay-Regularizer-for-Keras/EigenvalueDecay.py", line 57, in *call* domin_eigenvect = K.dot(WW, o) File "/home/andrewcz/miniconda3/lib/python3.5/site-packages/ keras/backend/tensorflow_backend.py", line 340, in dot if ndim(x) is not None and (ndim(x) > 2 or ndim(y) > 2): File "/home/andrewcz/miniconda3/lib/python3.5/site-packages/ keras/backend/tensorflow_backend.py", line 220, in ndim dims = x.get_shape()._dims AttributeError: 'numpy.ndarray' object has no attribute 'get_shape' help please. Many thanks, Andrew — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#2>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ARcw09IMom1499AmJoE9TjsDB_LLZ1rsks5rzsdLgaJpZM4NIWNr> .

[avinashkd44]

pip install git+git://github.com/avinashkd44/keras.git --upgrade # You will get here eigenvaluedecay regularizer for tensorflow backend.

Hi Andrew and Oswaldo
I have tested and modified your file and now its working for Tensor flow for mine.