Hi,I recently build a convolutional neural network,however, an error occurred in Keras deep learning network. I can successfully run the example on KNIME workflow(Train mnist classifier). So I do not know how to resolve my workflow.
Hi @Lemon and welcome to the forum.
What is the error message displayed in the console, or when you mouseover the red X on the Learner node?
The console said:
WARN Keras Network Learner 3:93 E:\ANACONDA-python\envs\py3_knime_dl\lib\site-packages\keras\engine\saving.py:292: UserWarning: No training configuration found in save file: the model was not compiled. Compile it manually.
ERROR Keras Network Learner 3:93 Transmitting training data to Python failed.
ERROR Keras Network Learner 3:93 Execute failed: An error occured during training of the Keras deep learning network. See log for details.
Hi,if you have any idea please tell me soon, thank you!
I’m a bit baffled by the error messages here, not having seen them before. I gather you were able to run the standard MNIST workflow, but you are now having problems after you made some changes to the structure and/or configuration of the layers, is that correct?
Could you export and upload your actual workflow instead of just a screenshot? That might help.
Yes,I only add two “String Manipulation” in the workflow.
My workflow is as followed.
The size of 600 pictures is 6804804. Is it too big to execute?
I can‘t upload the data,cause the workflow uploaded in KNIME hub limits 50MB. Or I should upload some files? But I don’t know how to do.
That could be the issue. If I recall, the size of the MNIST digits is quite small, something like 25x25. The Keras Input Layer explicitly defines the size, and yours are much larger of course.
Thanks for your attention, I have reduced the size of my images by using “image resize”. And the issues have been solved( [Transmitting training data to Python failed.]).
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.