-
Hi,I have been reading the paper and looking at the code, and I don't understand how the KL is being calculated in the loss term.
The loss calculation has a KL term with flow:
`(q_z.log_prob(z0).s…
-
Hello! I'm working on a project (to be open-sourced) that could use a few Autograd normalizing flow implementations, and I was wondering if you could add a license? It would save me a bit of time if I…
-
Hello,
Could I ask why you normalize the flow by
vgrid[:,0,:,:] = 2.0*vgrid[:,0,:,:].clone() / max(W-1,1)-1.0 in the warping function?
Due to minus value in flow, the vgrid could also have minus …
ghost updated
5 years ago
-
It will be useful to add [BatchNorm Transform](https://www.tensorflow.org/api_docs/python/tf/contrib/distributions/bijectors/BatchNormalization) to stabilize the training in the normalizing flows, but…
-
(possibly duplicate)
If two revisions occur at the same `now`, `%gall` will send duplicate "build $app now" requests to ford, expecting the second to produce an updated(different) version. This cause…
-
Hi,
I am getting different versions when running:
* an incorrect version using a linux docker image in version 5.0.0 or 5.0.1 with my repo mounted
* a correct version using the gitversion binary …
-
It may be far too late in the release cycle for this, but in writing an outline for a presentation on RxJava 2 for the last month I've come to think we're erroneously shipping two libraries as one.
R…
-
**Describe the problem**
When importing Bro/Zeek/Corelight logs geo / geo_point is not being populated.
Filebeat with Zeek module enabled > Logstash > Elasticsearch parses the log and loads it but…
-
Where is the batch normalization implementation for Multi-GPU scenarios? How does one keep track of `mean`, `variance`, `offset` and `scale` in the context of the Multi-GPU example as given in the [CI…
-
Add proper support for the DICOM transmission protocol.
Imported Issue. Original Details:
Reporter: chrisl
Created: 2007-10-30T13:35:10.000-0700