hasan7n
hasan7n
Suggestion from the medperf team: The effect of the `--mount` arg is only for input volumes. The name could be misleading as it can be interpreted to act on all...
Running `mlcube configure` with an MLCube that uses a local docker image that hasn't been pushed yet to the remote registry is throwing an error (and exiting with status 1)....
When using the FeTS tool ([script](https://github.com/aristizabal95/FeTS-AI-Front-End/blob/1ef893952116690dfe49a6262622b1a07b31220e/src/applications/PrepareDataset.py)), we noticed two sources of memory build-up: 1. Memory build-up due to figure_generator ([lines](https://github.com/aristizabal95/FeTS-AI-Front-End/blob/1ef893952116690dfe49a6262622b1a07b31220e/src/applications/PrepareDataset.py#L145-L152)), around 200MB per call. 2. Memory build-up due to GaNDLF's...
**Is your feature request related to a problem? Please describe.** The STDOUT can be full of warnings, it should be more user friendly **Describe the solution you'd like** Make GaNLDF's...
This PR adds the possibility of having multiple collaborator nodes use the same certificate for communication with the aggregator. The goal is to separate the dataset "identity" from its owner...
Refactor entities and use inheritance to not repeat code. This is needed since for FL support we are adding 4 new entities with similar interface. This PR also removes the...
More information is needed, but it seems the filtering logic (at the comms layer, or at the entity layer) doesn't take into account models associated with multiple benchmarks.
Currently our integration tests and server unit tests assume a sqlite3 database, while our production database is PostgreSQL. We should use PostgreSQL in testing since 1) some sql capabilities are...
Some users are experiencing an "Unknown or Invalid refresh token" error. This may be caused when a user leaves the MedPerf client for a month, since the refresh token lifetime...