-
**Describe the bug**
When values are extracted from a data frame with a multi-index, the key is supposed to be an accepted tuple, but it is not.
**To Reproduce**
1. Provide a minimal runnable `pa…
-
Hello, we are using s3fs to store data on s3 as part of our AI evaluations library [Inspect AI](https://github.com/UKGovernmentBEIS/inspect_ai), and we are seeing that using multi-processing (but inte…
-
Spent a while tracking down an issue where the number of HMC transforms mismatched the number of volumes in a BOLD series. Turned out is was multi-echo data where echos 1 & 2 had n volumes, and the th…
-
### Problem to solve
This was a feature in the V2 datatable via #9958 that I can't seem to find in V3.
When using "show-select"s in a VDataTable there is currently no native way of quickly selecti…
-
This is of course quite subjective but I feel like two thirds of my "incidents" are not "metric x has exceeded threshold y" but simply "service x has stopped reporting any metrics".
I realize that …
-
It seems that audio data is forced to be 1 or 2 channels, isn't it?
Is there a reason for this? Is there a way to simply enable multi-channel (>2) data support?
-
Thanks for your excellent work!
The provided dataset includes point cloud data and four cameras' intrinsic/extrinsic parameters. Could you please guide me on how to generate RGBD images for each ca…
-
### Multi-node TPU Training with JAX
The [multi-GPU JAX training guide](https://keras.io/guides/distributed_training_with_jax/) is helpful, but it's unclear how to extend this to multi-node TPU set…
-
Hello,
I am interested in using your model for a project focused on multi-channel source separation. I was wondering if you could provide any guidance, best practices, or documentation that would …
-
Right now we are using (a subset) of Libri Lite which is a very big (60k hours) dataset of audiobooks read by thousands of speakers. It is pretty good but there is a lot of (probably more expressive a…