-
## Description
In tutorial0 (or wherever Plugins is used) I encounter the following error:
A non-annotated attribute was detected: `protected_cols = ['seq_id']`. All model fields require a type anno…
-
Hi there!
In the 60-minute blitz tutorial (https://fluxml.ai/tutorials/2020/09/15/deep-learning-flux.html), the part where we train a network on CIFAR10 takes longer than expected. Could it be becaus…
-
### 🐛 Describe the bug
```code
model_type FLOW
clip missing: ['text_projection.weight']
Requested to load FluxClipModel_
Loading 1 new model
Requested to load Flux
Loading 1 new model
0%| …
-
This happens occasionally. When this happens the actual audio output is affected too. E.g. wsjt-x would show a constant high audio input level. Restarting the app or changing the sampling rate from t…
-
The algorithm described in [Extreme Learning](http://extremelearning.com.au/an-improved-version-of-bridsons-algorithm-n-for-poisson-disc-sampling/) is incorrect: it does not generate blue noise.
Bl…
-
-
Hello,
I wanted to ask something regarding the Joint-training based methods in the ACD and Hold-Out setting. Is it correct that the current training scripts always apply the meta-learning approac…
-
I think there is a bug in the way formatting is handled for Llama models [here](https://github.com/ShishirPatil/gorilla/blob/main/berkeley-function-call-leaderboard/model_handler/llama_handler.py#L13)…
-
This is an FYI / guidance for those who are currently running `cmdstanpy>=1.0.2`, specifically when doing full MCMC sampling (`mcmc_samples > 0`). We've noticed that `cmdstanpy` now logs warnings fair…
-
Hello, thanks for the nice package.
I was working on an application where I wanted perfect prediction in a classification task and found that I was unable to do that with `partial_frac = 1.0`, whic…