Comments (4)
When predicting we have to enter certain 'forcing' values, if those values are t, t+1, t+2... hours' values instead of t, t+6, t+12.... wouldn't that generate hourly values.
from graphcast.
An hourly model would indeed be very interesting for several use-cases. Assuming this model would be used for more short-term predicitons (0-48 hours ahead) it would further be great to optimize the forecast accuracy for 1 day lead-times instead of the 3.5 day lead-time it is currently optimized for.
@abhinavyesss : I think in order to produce meaningful hourly forecasts the model would have to be retrained on hourly ERA5 data instead of just adjusting the 'forcing' values.
@erinboyle : note that for commercial use cases like the energy market there would still be licensing issues since the model weights are published under the non-commercial CC BY-NC-SA 4.0 license.
from graphcast.
The challenge with training a 1 hour model is that we don't have good 1 hour data. ERA5 does indeed have 1 hour data, but it only incorporates new observations every 12 hours, so the other 11 hours are just predictions from the 2016 HRES model. This makes a model trained on 6 hour intervals have a weird learning problem where half the time it's learning real physics and half the time it's learning NWP model physics. This is somewhat (mainly?) mitigated by fine-tuning on the HRES dataset that incorporates real observations every 6 hours. Unfortunately we don't have any dataset that incorporates observations every 1 hour. It's certainly possible to train a model that trains on the 1 hour ERA5 data, but I wouldn't expect it to outperform HRES for the first 12 hours. It may outperform HRES when taking a better graphcast prediction as its input, but that is likely also out of distribution (due to blurring), so even that is not guaranteed. Either way, I agree that this would be a useful thing to have, but has a fair amount of challenges to do well.
from graphcast.
My guess is that MetNet-3 is more suitable for higher time resolution. Unfortunately, there are no official plans to open source MetNet-3.
- Lucidrains has an open-source reproduction of the code (no weights yet) here: https://github.com/lucidrains/metnet3-pytorch
- Open Climate Fix replicated the training of MetNet-1 and 2 and released their pre-trained weights, as shown here: https://github.com/openclimatefix/metnet
from graphcast.
Related Issues (20)
- Jax Error only when TPU-enabled runtime selected HOT 2
- Predicting Forecast for 10 Days , 5 Days HOT 2
- Obtaining successive forecasts based on previous predictions HOT 2
- Are forcing variables repeated? HOT 2
- Haiku needs all `hk.Module` must be initialized inside an `hk.transform` HOT 1
- About loss weights HOT 3
- GPU / TPU memory requirements for training HOT 3
- [GraphCast Operational Model] Issue with Negative Precipitation Data in GraphCast Operational Model Output HOT 2
- How to get the data in the paper? HOT 1
- weights license - use of graphcast HOT 5
- Graphcast error on Mac os HOT 1
- Problems feeding data to operational model: Target variable geopotential_at_surface must be time-dependent HOT 1
- when is the prediction result of this demo? HOT 2
- Forecasting beyond 10 days HOT 8
- Cyclone tracking
- There are some questions about forecasting. HOT 1
- Fine-Tuning Strategy for the GraphCast Operational Model HOT 2
- About the atmospheric variable “Vertical velocity”
- about the autoregressive finetuning HOT 1
- How to train a model by myself HOT 6
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from graphcast.