validation steps
See original GitHub issueHi, thank you for your work! It helps me a lot.
I run into a issue after I convert the brats data and start the training by : python train_isensee2017.py or python train.py, but it show me some problem with validation steps. Any suggestions about my issue?
(tensorflow)root@ubuntu:~/3DUnetCNN/brats# python train_isensee2017.py
Using TensorFlow backend.
2018-02-23 09:35:45.097413: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX
Loading previous validation split…
Number of training steps: 0
Number of validation steps: 0
Traceback (most recent call last):
File “train_isensee2017.py”, line 117, in <module>
main(overwrite=config[“overwrite”])
File “train_isensee2017.py”, line 112, in main
n_epochs=config[“n_epochs”])
File “/root/3DUnetCNN/unet3d/training.py”, line 88, in train_model
early_stopping_patience=early_stopping_patience))
File “/root/tensorflow/lib/python3.4/site-packages/keras/legacy/interfaces.py”, line 91, in wrapper
return func(*args, **kwargs)
File “/root/tensorflow/lib/python3.4/site-packages/keras/engine/training.py”, line 2115, in fit_generator
raise ValueError(‘validation_steps=None
is only valid for a’
ValueError: validation_steps=None
is only valid for a generator based on the keras.utils.Sequence
class. Please specify validation_steps
or use the keras.utils.Sequence
class.
Closing remaining open files:/root/3DUnetCNN/brats/brats_data.h5…done
Thank you!
Issue Analytics
- State:
- Created 6 years ago
- Comments:13 (2 by maintainers)
Top GitHub Comments
File train.py/ train_isensee2017.py line No.46
set, config[“overwrite”] = True (False by default) " (original) config[“overwrite”] = False # If True, will previous files. If False, will use previously written files. "
Now working fine
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. If you have questions, feel free to join the Slack group or email me at davidgellis2@gmail.com.