Closed LBlend closed 3 years ago
I have verified that this is a compatibility issue between packages and not a fault on our end. We should specify the version number for python packages in order to prevent this from happening in the future.
Because of this, I have resorted to running the 1.0.0 release in production. I have to use a release without Tensorflow
Yep. I actually stumbled upon this problem in another project as well. It's most likely caused by having a too recent version of numpy. I'd suggest making sure to install a tensorflow-compatible numpy version. I fixed it by installing numpy==1.19.5
Yep. I actually stumbled upon this problem in another project as well. It's most likely caused by having a too recent version of numpy. I'd suggest making sure to install a tensorflow-compatible numpy version. I fixed it by installing numpy==1.19.5
This resolves issue. Great! 👌
I will go ahead and push the changes
What is wrong?
When trying to run the API, Tensorflow raises a
NotImplementedError: Cannot convert a symbolic Tensor (bidirectional/forward_lstm/strided_slice:0) to a numpy array. This error may indicate that you're trying to pass a Tensor to a NumPy call, which is not supported
System info