FengWuW2S
Members
This team has chosen to keep its participants anonymous.
Models
Model name
FengWu
Number of individuals supporting model development:
1-5
Maximum number of Central Processing Units (CPUs) supporting model development or forecast production:
48-1,000
Maximum number of Graphics Processing Units (GPUs) supporting model development or forecast production:
4-16
How would you best classify the IT system used for model development or forecast production:
Single node system
Model summary questionnaire for model FengWu
Please note that the list below shows all questionnaires submitted for this model.
They are displayed from the most recent to the earliest, covering each 13-week competition period in which the team competed with this model.
Which of the following descriptions best represent the overarching design of your forecasting model?
- Machine learning-based weather prediction.
What techniques did you use to initialise your model? (For example: data sources and processing of initial conditions)
The ERA5 dataset, includes temperature, pressure, humidity, and wind data at 13 pressure levels, and 6 surface data with 6-hourly forecasts. The initial field is delayed by one week. For example, the results submitted on November 9th comes from the ERA5 dataset dated November 2nd.
If any, what data does your model rely on for real-time forecasting purposes?
ERA5 dataset
What types of datasets were used for model training? (For example: observational datasets, reanalysis data, NWP outputs or satellite data)
ERA5 dataset
Please provide an overview of your final ML/AI model architecture (For example: key design features, specific algorithms or frameworks used, and any pre- or post-processing steps)
Based on SwinTransformer and utilizing VAE to perturb latent space features
Have you published or presented any work related to this forecasting model? If yes, could you share references or links?
None
Before submitting your forecasts to the AI Weather Quest, did you validate your model against observational or independent datasets? If so, how?
only use the real-time data to validate
Did you face any challenges during model development, and how did you address them?
How can we ensure that the quintile distribution of the predicted results matches the quintile distribution of the actual predicted results?
Post-processing can be used for adjustment.
Are there any limitations to your current model that you aim to address in future iterations?
real time forecast, use the analysis replace the era5 dataset
Are there any other AI/ML model components or innovations that you wish to highlight?
NO
Who contributed to the development of this model? Please list all individuals who contributed to this model, along with their specific roles (e.g., data preparation, model architecture, model validation, etc) to acknowledge individual contributions.
This team has chosen to keep its participants anonymous.
Model name
FengWu2
Number of individuals supporting model development:
1-5
Maximum number of Central Processing Units (CPUs) supporting model development or forecast production:
48-1,000
Maximum number of Graphics Processing Units (GPUs) supporting model development or forecast production:
4-16
How would you best classify the IT system used for model development or forecast production:
Single node system
Model summary questionnaire for model FengWu2
Please note that the list below shows all questionnaires submitted for this model.
They are displayed from the most recent to the earliest, covering each 13-week competition period in which the team competed with this model.
Which of the following descriptions best represent the overarching design of your forecasting model?
- Machine learning-based weather prediction.
What techniques did you use to initialise your model? (For example: data sources and processing of initial conditions)
weekly era5 dataset
If any, what data does your model rely on for real-time forecasting purposes?
only era5 dataset
What types of datasets were used for model training? (For example: observational datasets, reanalysis data, NWP outputs or satellite data)
era5 weekly dataset
Please provide an overview of your final ML/AI model architecture (For example: key design features, specific algorithms or frameworks used, and any pre- or post-processing steps)
Rolling forecast based on weekly average results
Have you published or presented any work related to this forecasting model? If yes, could you share references or links?
None
Before submitting your forecasts to the AI Weather Quest, did you validate your model against observational or independent datasets? If so, how?
The results from the competition were verified both online and offline, and the results were largely consistent.
Did you face any challenges during model development, and how did you address them?
Data overlap makes real-time predictions difficult for the model, with a one-week delay. However, without overlap, the skill level drops rapidly due to insufficient data resolution.
Are there any limitations to your current model that you aim to address in future iterations?
Data overlap makes real-time predictions difficult for the model, with a one-week delay. However, without overlap, the skill level drops rapidly due to insufficient data resolution.
test in real-time analysis or correction
Are there any other AI/ML model components or innovations that you wish to highlight?
None
Who contributed to the development of this model? Please list all individuals who contributed to this model, along with their specific roles (e.g., data preparation, model architecture, model validation, etc) to acknowledge individual contributions.
This team has chosen to keep its participants anonymous.
Model name
FengWu3
Number of individuals supporting model development:
1-5
Maximum number of Central Processing Units (CPUs) supporting model development or forecast production:
48-1,000
Maximum number of Graphics Processing Units (GPUs) supporting model development or forecast production:
4-16
How would you best classify the IT system used for model development or forecast production:
Single node system
Model summary questionnaire for model FengWu3
Please note that the list below shows all questionnaires submitted for this model.
They are displayed from the most recent to the earliest, covering each 13-week competition period in which the team competed with this model.
Which of the following descriptions best represent the overarching design of your forecasting model?
- Machine learning-based weather prediction.
What techniques did you use to initialise your model? (For example: data sources and processing of initial conditions)
Did you face any challenges during model development, and how did you address them?
Data overlap makes real-time predictions difficult for the model, with a one-week delay. However, without overlap, the skill level drops rapidly due to insufficient data resolution.
Are there any limitations to your current model that you aim to address in future iterations?
Data overlap makes real-time predictions difficult for the model, with a one-week delay. However, without overlap, the skill level drops rapidly due to insufficient data resolution. test in real-time analysis or correction
Are there any other AI/ML model components or innovations that you wish to highlight?
None
Who contributed to the development of this model? Please list all individuals who contributed to this model, along with their specific roles (e.g., data preparation, model architecture, model validation, etc) to acknowledge individual contributions.
Dr. Tao Han and Dr. Xie
If any, what data does your model rely on for real-time forecasting purposes?
only era5 dataset
What types of datasets were used for model training? (For example: observational datasets, reanalysis data, NWP outputs or satellite data)
era5 daily dataset
Please provide an overview of your final ML/AI model architecture (For example: key design features, specific algorithms or frameworks used, and any pre- or post-processing steps)
Rolling forecast based on daily average results
Have you published or presented any work related to this forecasting model? If yes, could you share references or links?
None
Before submitting your forecasts to the AI Weather Quest, did you validate your model against observational or independent datasets? If so, how?
The results from the competition were verified both online and offline, and the results were largely consistent.
Did you face any challenges during model development, and how did you address them?
None
Are there any limitations to your current model that you aim to address in future iterations?
Real time forecasting and the data range shift by long-term rollout
test in real-time analysis or correction
Are there any other AI/ML model components or innovations that you wish to highlight?
None
Who contributed to the development of this model? Please list all individuals who contributed to this model, along with their specific roles (e.g., data preparation, model architecture, model validation, etc) to acknowledge individual contributions.
This team has chosen to keep its participants anonymous.
Submitted forecast data in previous period(s)
Please note: Submitted forecast data is only publicly available once the evaluation of a full competitive period has been completed. See the competition's full detailed schedule with submitted data publication dates for each period here.
Access forecasts data