Open amalmathews opened 4 years ago
This repo is not updated for tf 2.x
. You should downgrade tf 1.13.x
But then how to use GPU for computing GPU is only available for TensorFlow 2.x, with CPU computing take far more time
You can use GPU for computing as well, there will be no problem if you downgrade tf 1.x
. The main thing you need to know and consider now is that there is no official support (from this Github author) for tf 2.x. Though some people are trying to make it work, but there may be still a lack of something (I didn't try on their approach).
And just to inform you, I've recently made a notebook on the on-going kaggle competition about wheat head detection using this implementation. The (by default or updated kernel) used tf 2.1
so I've faced some trouble with that and at the end manage to downgrade tf 1.x
. While facing this issue, I've made a request in the discussion forum, link here, maybe you can find some more insight on this.
You can use GPU for computing as well, there will be no problem if you downgrade
tf 1.x
. The main thing you need to know and consider now is that there is no official support (from this Github author) for tf 2.x. Though some people are trying to make it work, but there may be still a lack of something (I didn't try on their approach).And just to inform you, I've recently made a notebook on the on-going kaggle competition about wheat head detection using this implementation. The (by default or updated kernel) used
tf 2.1
so I've faced some trouble with that and at the end manage to downgradetf 1.x
. While facing this issue, I've made a request in the discussion forum, link here, maybe you can find some more insight on this.
I did but now I'm getting a new error
OperatorNotAllowedInGraphError: using a `tf.Tensor` as a Python `bool` is not allowed in Graph execution. Use Eager execution or decorate this function with @tf.function.
Please help with this recurring error
What is your tf
version now? Where you're running your code? Can you share the reproducible code of yours?
I have recently updated my TensorFlow to 2.2 for working with GPU support but after upgrading my maskRCNN program start showing many kinds of errors.
As per my little experience with deep learning, I solved most issues bu is stuck with this one error which I'm getting and is not able to solve it.
I have been such with the problem and I really need help. Someone please help me with the issue