Comments (3)
Thanks for the report and clear example. It looks like it's because the variables were created with tf.Variable rather than tf.get_variable, this is known to interact poorly with variable scopes. It's mentioned in the docs but might be a bit buried.
Here's my code:
https://gist.github.com/tfgg/d815ec514697a16bc1067acdbb6f0e1f
which outputs
[u'main_module/submodule_a/a:0', u'main_module/submodule_b/b:0']
[u'main_module/submodule_a/a:0', u'main_module/submodule_b/b:0']
from sonnet.
@tfgg Thanks for the clarification and the example! I must have missed that bit in the docs. I'm not creating any variables directly so it's probably down to some TF slim utils.
Besides being kind of annoying, does this behaviour cause any unwanted effects on the computation?
Closing the issue, thanks!.
from sonnet.
I'm not 100% sure, but I think variable sharing would fail (since it's creating a new variable each time), but I'd want to check that first :)
It's certainly possible that something in TF-slim caused this, it might not always interact well with Sonnet, though I think that also uses tf.get_variable.
from sonnet.
Related Issues (20)
- Understanding how to use mixed precision training HOT 2
- Conv2D initialize error in tf function
- Typo HOT 1
- this portion of attention code looks incorrect
- Cannot install sonnet on mac m1 HOT 5
- Decaying learning rate HOT 1
- Can you explain the logic of updated_ema_cluster_size in VectorQuantizerEMA?
- Convert a Sonnet model to a Keras model HOT 1
- (Maybe) in-consistency between VQ-VAE paper and its implementation. HOT 2
- Regarding LearningRateSchedule for ADAM optimizer HOT 2
- Learning rate and optimizer used for the VQ-VAE-2 paper HOT 2
- I have a question about the perplexity term (in the VQ-VAE). HOT 2
- Issues using Sonnet HOT 3
- How to save the checkpoint for self-defined model?
- TensorArray Not Used on line 71 of sgd_test.py HOT 2
- Is there a stereotypical way to visualize the network strucuture when using Sonnet? HOT 1
- sonnet abstract module in tensorflow2 HOT 10
- Laplace smoothing for EMA codebook update
- Basic import fails on Colab: cannot import name 'context' from 'tensorflow.python' HOT 11
- AttributeError: module 'nest' has no attribute 'Create' HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from sonnet.