Jacob Prince-Bieker
Jacob Prince-Bieker
# Pull Request ## Description Adds the same pydocstyle linter as in https://github.com/openclimatefix/nowcasting_utils/pull/23 Relates to issue #1 ## How Has This Been Tested? It runs in GitHub Actions - [...
## Detailed Description https://arxiv.org/abs/2112.13776 This paper tries to get uncertainty estimation by using stochastic attention, so a slightly different way than the skillful nowcasting one. I've only very quickly skimmed...
https://vision.eng.au.dk/cloudcast-dataset/ seems to be EUMETSAT data with type of clouds labelled for a few years of data. Seems like its the Cloud Analysis data, which is what I've been waiting...
## Detailed Description To know how changes are affecting model performance, it could be nice to be able to run a model with commands from PRs and record how it...
## Detailed Description https://www.iarai.ac.at/weather4cast/ (on GitHub here: https://github.com/iarai/weather4cast) has some nice weather data from around the world, including cloud masks, all at 4km resolution, so similar to what EUMETSAT gives....
Atmospheric Motion Vectors for example could be useful to include where winds seem to go? Still need to look into it more, but there are a few other rapid scan...
Various ideas Some from @JackKelly: - [ ] Pretrain predicting next frame from past two - [ ] Simulated clouds/optical flow - [ ] Try [AutoFlow](https://people.csail.mit.edu/celiu/pdfs/CVPR21_AutoFlow.pdf) like in Perceiver paper
## Detailed Description Similar to Perceiver IO (they even reference it in the blog post) this is a modality agnostic model, with a public PyTorch implementation, no weights for vision...
## Detailed Description https://medium.com/syncedreview/deepminds-retro-retrieval-enhanced-transformer-retrieves-from-trillions-of-tokens-achieving-c54313fec51d ## Context Paper is mostly about language models, but has an interesting idea for searching for similar text to help the model learn, and seems to...
## Detailed Description https://arxiv.org/abs/2111.12993 Seems similar to Perceiver, and they mention Perceiver as a related model, but their training is a bit different. But it is still a multi modal...