-
as noted in #68, we could do with a few constants for the sizes of some things. Those constants could then be device-specific, or maybe even calculated at startup (if we define them the right way, no …
-
Hi,
Love this, neat, simple and the default style is perfect for a dark dashboard!
One request - it's hard to get it to show floating point values in the indicator at the moment, because even if…
-
not sure this is a good idea, but I got the urge to try scrolling over just now and felt the lack. It seems like it would be possible to create a gene list based on the columns of the alignment, at le…
-
i just updated last night to the new version and now the picture of rares items are just beyond fudge up
![image](https://user-images.githubusercontent.com/46448848/50771830-c94a7480-1259-11e9-9f9a…
-
We currently "fudge" autograd.Function by running through the forward as if it was the function and rely on the differentiation of that to work.
(This is not good when there is `.detach()` or some su…
t-vi updated
2 months ago
-
Even after 15% fudging, still off compared to a wahoo cadence sensor.
One thing to try is taking the entire gyro vector, not just gyro_z.
jdt23 updated
3 years ago
-
Here are some ideas for additional diagnostics inspired by discussions with @jonathanlilly and @JeffreyEarly.
### Eigensolver for vertical modes
I think it would be nice to have an eigensolver t…
-
When building example with ``-DPICO_PLATFORM=rp2350-arm-s`` it stucks on first 8-bit writes.
-
### 🐛 Describe the bug
The current method to estimate the numerical errors in test_transformers.py is based on the difference b/w results from current precision and higher precision tensors
https:…
-
Actually, not by expected time (2s) plus a nearly 2s of fudge. Is it even wired up?
See conversations here: #252