An implementation of Perlin noise in PyTorch has been made by the X user honualx.
Unresolvable The public record could not settle it.
Factual: Statements about the present or the past, checked against the record: did they hold up?
What would count as true or false
True if a Perlin noise implementation in PyTorch existed by 2020-09-30 and is credited to the account @honualx (for example a public GitHub repository or gist by the same person, linked from or attributed to that X account). Mostly true if the implementation is by honualx but is of a closely related noise, such as fractal Perlin, or a contributor is credited alongside. Mostly false if it existed but was mainly by someone else or not in PyTorch. False if no such implementation by honualx existed by that date, as shown by the account or repository history.
Set before the research began.
Evidence (5)
-
E1 GitHub - duchesneaumathieu/pyperlin: GPU accelerated Perlin Noise in python GitHub · 19 February 2020
“Using pytoch as the array backend makes the GPU acceleration trivial. Also it uses batch sampling to better use parallelization.”
-
E2 Perlin noise in PyTorch GitHub Gist
“# ported from https://github.com/pvigier/perlin-numpy/blob/master/perlin2d.py”
-
E3 pytorch-perlin-noise/perlin_noise/perlin_noise.py at main · tasptz/pytorch-perlin-noise GitHub · 29 April 2023
“meta-description: Simple perlin noise with pytorch tensors. Contribute to tasptz/pytorch-perlin-noise development by creating an account on GitHub.”
-
E4 Perlin noise.ipynb GitHub Gist (adefossez) · 1 October 2020
“out = perlin_ms([rho**-i for i in range(4)], 6, 6).cpu().numpy()”
-
E5 pyperlin PyPI · 23 October 2020
“Using pytoch as the array backend makes the GPU acceleration trivial.”
Reasoning
E4 is a 'Perlin noise' notebook gist by the GitHub user adefossez, dated 2020-10-01. Its quoted code calls a multi-scale perlin function and then .cpu().numpy(), which suggests a PyTorch implementation. However, no item links adefossez to the X account @honualx, and the gist's date falls one day after the claim date. The other items (E1, E2, E3, E5) are Perlin-in-PyTorch implementations by other authors and do not mention honualx. Because nothing shows a PyTorch Perlin implementation credited to @honualx by 2020-09-30, the evidence does not settle the claim.
Cite this claim
Yann LeCun, “An implementation of Perlin noise in PyTorch has been made by the X user honualx.” (said 30 September 2020). Verdict: unresolvable. Reliability API, claim c_108a0b546322, method 0.1.0+c876030c, dataset release v1.1.2. https://epistacklabs.com/claims/c_108a0b546322
This claim is in dataset release v1.1.2, with the same verdict.
Reply to this claim
More factual claims by Yann LeCun
-
Mostly true
“This white-noise sounding signal with a flat spectrum is the pseudo-random bit sequence used to pre-train the adaptive equalizer that cleans up the mushy analog signal into clean bits.”
-
Resolved true
“Opacus: a new PyTorch library for ML with differential privacy.”
Facebook AI released Opacus, a new PyTorch library for machine learning with differential privacy.
-
Resolved true
“Announcing PyTorch Hub!”
-
Mostly true
“We are using Torch7 for many projects (as does Deep Mind and several groups at Google) and will be contributing to the public version.”
Facebook AI Research uses Torch7 for many projects, as do Deep Mind and several groups at Google.
-
Resolved false
““Those agents are doing exactly what they’ve been asked to do,” LeCun said.”
Verdicts are made automatically; how verdicts are made.