-
-
Thanks for sharing your amazing notes! In `/seminars/2018-10-Normalizing-Flows-NICE-RealNVP-GLOW/2018.10.10_Normalizing_Flows_NICE_RealNVP_GLOW.pdf` page 9, the expression for the normalizing flow has…
-
I haven't run comparison with original code, but does have actNorm layer in the coupling layer improve performance ?
https://github.com/kamenbliznashki/normalizing_flows/blob/71e630ced23c33ff34db84…
-
--------------------------------------------------------------------------------------------------
### Issue Description
Provide a brief description of the issue.
I think the ConditionedPlanarF…
-
Hi @fehiepsi, following the [discussion](https://forum.pyro.ai/t/using-pyro-inverse-autoregressive-flow-in-hmc-inference/1524) on the forum.
### Issue Description
Issue appears when I try to use N…
-
## Task
#### Description
Ingestion engines need to be able to insert time to payloads.
There is some discussion at #42 around handling the concept of time, and this can be done in parallel with…
-
## 🐛 Bug
I think the Kl divergence is summing over the batch dimensions when it shouldn't, at least for the Gaussian case
Here's the transformed distribution function:
```
def _kl_transformed_tr…
-
Hi Krzysztof,
I'm studying Glow with your code and being confused about y and z, please let me ask some questions.
**Q.1**
My understanding is as follows:
- y and z come from multiscale arch…
-
This is a copy of an issue I posted for pytorch: https://github.com/pytorch/pytorch/issues/29698
I think this is an issue with the _kl_independent_independent class which is summing over batches wh…
-
### Issue Description (Feature request)
I was going through the implementation for the PlanarFlows Module because it was bothering me how exactly they worked. I think the way they're set-up right no…