Addison Klinke

A few thoughts here 1. What about supporting inequality operators in addition to arithmetic? 2. I like the `${eval:}` or `#{expression}` syntax mentioned a lot. That seems to be very...

Also affecting `api/storages/s3`. @makseq is there a recommended workaround for the time being? I need to be able to programmatically sync storage (which requires the storage ID that I'm trying...

Here's a very minimal example which I hope helps in pinpointing the error. It's a single layer model with only one standard PyTorch convolution. My hunch is that the input...

Figured out my issue (illustrated below with a conv layer). Unfortunately it's not the same as Detectron because the code by @ptmdmusique already uses the correct `inputs_single` approach ```python import...

@TobyRoseman regarding my second post - the reason I got an error from `ct.convert` was that I was *inadvertently* using a different model input than what had been used for...

Overall, I strongly support some kind of pipeline execution native to MLflow. I think that would make it a much more complete platform for MLops. For some background on my...

@zbloss Agreed, I think there's already a plethora of pipeline tools out there (Pachyderm, ClearML, MLRun, ZenML, Flyte, Kedro... the list goes on) so it would be nice to start...

I'd find this super useful for linking to a specific git commit or Docker image within a remote repository. It'd be even better if MLflow recognized commit and image hashes...

Any recent updates on the progress here? This would be a great feature! I also unfortunately do not know Java, so I cannot help in contributing a PR

@romesco Here's my approach for pure dataclasses. I'm not sure whether this would be considered best practice in Hydra - it sounds like typically `MyDatamoduleConf` would be a YAML file...