Jacob Prince-Bieker

Hi, there needs to `pip install ocf_blosc2` and then `import ocf_blosc2` for it to work.

Yes, it is! And yeah, happy to have you take it!

Hi, Sorry for the delayed response, I was on annual leave. I just pushed a minor update that should also fix some incompatibilities from using newer dependencies. For 2. you...

Hi, we used an A6000 to do the intial training of DGMR, which has 48GB of GPU memory. We haven't tried it with multiple GPUs yet, last time I did,...

Hi, the 48GB was enough for a few steps, not sure if all 6 though, its been awhile unfortunately. There were a few issues, originally it was because of `nan`...

Hi! So there are pretrained weights, but they are not very good. Part of the issue has been that the models take a huge amount of GPU memory at the...

Ah, good spot, thanks!

Hi, are you using multiple GPUs? By default the run.py tries to use 6 GPUs, although it should be changed to 1. The spectrally normalized layers in PyTorch don't seem...

Hi, A few questions: 1. Have you setup a virtual environment or conda environment? 2. Have you tried `pip install metnet`? 3. What does your data look like?

The data loading and format is more up to you. What MetNet takes is a series of multi-channel images, with the input being of the format [Batch, Channels, Timesteps, Width,...