-
JupterLab has started recently unconditionally adding `orig_nbformat: 1` into edited notebook since https://github.com/jupyterlab/jupyterlab/pull/10118. This broke notebook trust, since `orig_nbformat…
-
```python
import aesara.tensor as at
from etuples import etuplize
a = at.scalar('a')
etuplize(at.square(a))
# e(e(, sqr, ), a)
etuplize(at.pow(a, 2))
# e(e(, pow, ), a, TensorConstant{2})…
-
### Description
Found out by @tvwenger in https://github.com/pymc-devs/pymc/discussions/6905#discussioncomment-7027204
```python
import pymc as pm
from pymc.logprob import conditional_logp
…
-
To check that 30_000 triples are isomorphic to itself takes 12min even though parsing the file only takes 30ms (found by running https://github.com/jeswr/rdf-iso-test/blob/master/test.ts).
I apprec…
-
Making some notes from discussion had earlier today with @dlongley, where he stated something to this effect:
LD proofs allow data translation, remove redundant bloat, and keep the data model layer…
-
I found pynauty via https://stackoverflow.com/a/14574330 where somebody claims that any two isomorphic graphs yield the same certificate. Is this correct?
The inverse relation (any two graphs with …
pkel updated
2 months ago
-
```python
import aesara
import aesara.tensor as at
x, y = at.scalars("x", "y")
f = aesara.function([x, y], [x + y, x + y, y + x, y + x])
aesara.dprint(f)
```
```python
DeepCopyOp [id A] ''…
-
Although there is a [canonical version of N-Triples](https://www.w3.org/TR/n-triples/#canonical-ntriples), we will not find recommendations on the specific features of the N3.
Dependencies:
- N-Tr…
-
There's nearly enough in place to start automating the application of proximal methods. I've outlined some of the basic ideas [here](https://brandonwillard.github.io/a-role-for-symbolic-computation-i…
-
@whyrusleeping said (in the filecoin community slack)
---
@jbenet i’ll do a pass on canonicalization for the car format, but one thing i’m thinking through right now is that you can’t just naive…