Comments (6)
I tried to reproduce the error you are describing, but for me saving and loading of models with MADE nets works fine with model.save
and model.load
no matter if preprocessing
is a lambda function or not.
Can you provide a code snippet to reproduce this error, please?
Best regards,
Vincent
from normalizing-flows.
n_flos: int
dim: int
n_layers: int
hidden_units: int
for _ in range(n_flows):
flows += [
normflows.flows.AutoregressiveRationalQuadraticSpline(
dim,
n_layers,
hidden_units,
)
]
flows += [normflows.flows.LULinearPermute(dim)]
# Set base distribution
base = normflows.distributions.DiagGaussian(dim, trainable=False)
# Build flow model
nf = normflows.NormalizingFlow(base, flows)
# Save
torch.save(nf, 'model.ckpt`)
This would give me: Can't pickle MADE.<local>(lambda)...
from normalizing-flows.
thanks, I could reproduce the error now and replaced the lambda functions with something that can be pickled. Feel free to try it out yourself by installing the package from the current master branch.
Best regards,
Vincent
from normalizing-flows.
Thanks! :) Is this going to be pushed to PyPI, too?
from normalizing-flows.
sure, I will do so in about 1-2 weeks.
Best regards,
Vincent
from normalizing-flows.
I just released a new version of the package on PyPI. Hence, I'll close this issue.
from normalizing-flows.
Related Issues (20)
- Putting examples in the documentation HOT 5
- Forward and Inverse with log det function for `MultiscaleFlow`
- multi-gpu implementation HOT 1
- How the inverse was calculated HOT 1
- Conditional Flows implementation / documentation HOT 2
- Negative KL divergence HOT 3
- issue about ConditionalNormalizingFlow HOT 2
- The original glow seems to use `ConditionalDiagGaussian` HOT 1
- exp and sigmoid may cause inf. HOT 3
- Could you give an example for NICE? HOT 1
- NICE demo? HOT 1
- What dou you mean by "Augmented Normalizing Flow based on Real NVP"? HOT 1
- one-dimensional coupling flows do not work HOT 3
- Seeking Advice on Designing an Invertible Neural Network for Fission HOT 2
- Calculating forward KL divergence (probability density maximization), I get negative loss results on my dataset, is this reasonable? HOT 1
- Cannot have an odd latent_size (working with 2, 4, etc. , but not 3 or 5), shape problem HOT 2
- Conditional Coupling Layers
- Documentation clarification of forward_kld
- Generalization for higher-dimensional data
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 normalizing-flows.