keras
keras copied to clipboard
AttributeError: 'VocabWeightHandler' object has no attribute 'name'
System information.
- Have I written custom code (as opposed to using a stock example script provided in Keras): https://colab.research.google.com/drive/1BD9G1o7CNxAaKKyihCKx0wa1zW-tmn0r?usp=sharing
- OS Platform and Distribution (e.g., Linux Ubuntu 16.04):
- TensorFlow installed from (source or binary): pip
- TensorFlow version (use command below): 2.6.0
- Python version: 3.8
- Bazel version (if compiling from source):
- GPU model and memory:
- Exact command to reproduce:
Describe the problem. Tensorflow profiler crashes when using string categorical layer, not allowing to profile models with those layers.
Maybe the same issue
The same issue happened to me.
I "fixed" it by changing the Tensorboard(histogram_freq=0), but it seems you mentioned it's a bad idea as we can't get tensorflow profiler
in tensorboard.
This is a known bug with some special classes we introduced for handling StaticHashTable and MutableHashTable in keras layers (TrackableWeightHandler and VocabWeightHandler). Both the classes have caused a number of issues, so I believe the best approach here would be to remove them entirely. We will need to think about the design of how best to do that.
Is there any update on this issue?