-
In a project with 80 packages, `snapshot()` takes roughly two minutes on my machine. I think the reason is the recursive call to `getPackageRecords()` which seems to walk _all_ dependency trees for _e…
-
I'd like implement an additional API for "zipped" parallel computation with the SatrecArray. The [current computation](https://github.com/brandon-rhodes/python-sgp4/blob/master/extension/wrapper.cpp#L…
-
we are training on a wasserstein loss function
our network output is a PDF tensor of size 1xN**3, where N=10, etc, that is the pdf values for a 3D grid of NxNxN
For the true wasserstein distance…
-
Hi,
I have a code that samples a likelihood function using emcee in a multithreaded framework. the likelihood is computed by tinygp.
Now I have vectorized part of the likelihod function for runn…
-
Just a question / suggestion: this are the results of profiling:
```
ncalls tottime percall cumtime percall filename:lineno(function)
571 7110.513 12.453 7110.513 12.453 {built-i…
-
It is more efficience to calculate vectorized negative log-likelihoods than to loop through individual observations.
We may want to vectorize this in the future. It's redundant/inefficient to calc…
-
## 🐛 Bug
The output of `F.softplus` is different depending on the number of elements in the input tensor.
## To Reproduce
```
import torch
import torch.nn.functional as F
problem_value = -1.…
-
see: https://github.com/VictoriaMetrics/VictoriaMetrics/blob/74bb9ea734190d41397f8227d7dbc298ca49547e/lib/uint64set/uint64set.go#L764
if use 256 bit simd instruct, code will speed up 4 times.
an…
-
The `ColumnGeneratorCachedByIndex` is recommended for new cached column generators, but it can be significantly slower than the not-recommended way of first creating a ColumnGenerator and then adding …
-
Hi all, I have a batch of data and want to conduct the Kalman filter for each observation set. My data is `observations =(batch, num_time_lens, feature_dim)` so I use `tf.vectorized_map` for parallel …