-
I was looking at how sharing was added into equinox, and was a bit surprised at the approach taken. I had a similar challenge where I had a PyTree with shared jax.Arrays, and jit'ing a function that …
-
There has been increased interest from the community in using TVM for training. Relax, the next generation graph level IR of TVM, also faces the demand of training model.
We are building a trainin…
-
Hello,
I really like the idea to combine spacy with google stuff, thanks! I got some seemingly random deadlock problem...
I am basically doing n*(n-1)/2 comparison on 50K papers by their titles …
-
Hey @igul222, thanks again for your help with comments earlier.
I've been rereading both WGAN papers, and I understand that the critic loss is supposed to estimate the wasserstein distance. Below …
-
I am following a [tutorial](https://github.com/MyoHub/myosuite/blob/main/docs/source/tutorials/SAR/SAR_tutorial.ipynb) that trains the [myosuite myoHandReorient8-v0 environment](https://myosuite.readt…
-
Hello all,
I have a very simple question. How to compute gradients w.r.t. a subset of parameters. Let's say I have a model like:
```swift
struct Model: Layer {
var a = Tensor(1.0)
var…
-
# Summary
I propose to:
- Retroactively name the current (as of May 2021) IconVG wire format as FFV 0 (File Format Version 0).
- Consider FFV 0 a deprecated experiment and no longer supported.
…
-
The current HSV format for ColorEdit presented to the user is a byte integer (0-255) or float (0.0-1.0) however both Photoshop and Gimp use degrees for H (0.0-360.0) and % (0.0-100.0) for SV.
For e…
-
Thanks for your code.
It helps me understand the BiDAF in details.
However, I found the model had no performance increasing.
Every epoch, the metric is always the same.
And then, I found it's th…
-
## 🚀 Feature
I would like to request the addition of support for Recurrent Neural Networks (RNNs) in the KFACOptimizer class. Currently, the KFACOptimizer class works for linear and 2D convolution …