Robert Gibboni
Robert Gibboni
This commit makes a few changes that allow you to refresh the brain data using something like the following code... `client = cortex.webshow(data) # data is a cortex.Volume, client is...
When I clone the repo using `git clone --recursive https://github.com/drivendata/competition-winners.git .` (from [UPDATING.md](https://github.com/drivendataorg/competition-winners/blob/master/UPDATING.md)), I get the error: > Submodule path 'pri-matrix-factorization': checked out '5b9c6900b1d9ef38a181cfe292716cef3b96d4e1' > fatal: No url found for...
The training logs are currently only printed out to the console, but it would be useful to save them to a text file alongside the other model assets. Propose adding...
[`auto_lr_find` has some settings](https://pytorch-lightning.readthedocs.io/en/stable/advanced/lr_finder.html) that would be useful to expose in the config YAML, such as `min_lr`, `max_lr`. We can add a new config section `AutoLrFindConfig` that allows those options...
Most times, the user will want to view a subset of the entire dataset. Just spit-balling, but things like return all data points from a: - date or date range...
It would be great to be able to adjust preprocessing parameters in real-time. #20 takes a big step towards this by registering the pipeline to the redis database, from which...
Currently, the software has no concept of runs. Certain preprocessing steps carry states with them, e.g., a running mean that needs to reset at the end of each run. **Proposed...
I'm using the ubuntu-hpc 2204 x64 Gen 2 image on a Standard NC24ads A100 v4 VM. I train a vLLM model that uses NCCL and observe the following error: Error...