https://github.com/openclimatefix/pv-pseudo-experiments
Repo containing the training and experimentation code for PV MetNet and Pseudo-Irradiance models
https://github.com/openclimatefix/pv-pseudo-experiments
Last synced: about 1 year ago
JSON representation
Repo containing the training and experimentation code for PV MetNet and Pseudo-Irradiance models
- Host: GitHub
- URL: https://github.com/openclimatefix/pv-pseudo-experiments
- Owner: openclimatefix
- License: mit
- Created: 2023-03-27T09:29:12.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2023-09-14T09:56:20.000Z (over 2 years ago)
- Last Synced: 2024-04-14T11:50:43.044Z (about 2 years ago)
- Language: Python
- Size: 139 KB
- Stars: 2
- Watchers: 4
- Forks: 0
- Open Issues: 4
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# PV Site and Pseudo-Irradience Model Training and Experimentation Code
Template Repository for OCF Projects
## Usage
### Batch Creation
To create new batches, modify the `configs/datapipe/forecastor.yaml` file to point to the appropriate files
Also update `configs/batch.yaml` and `irradiance/batch_writing.py` to write to the correct places and with the correct options.
Then run `python run.py --config-name batch` to create the batches.
To change what data is being created with the datapipe, the changes have to happen to the `ocf_datapipes.training.pseudo_irradiance_datapipe`, which is what handles
the data processing.
### Inference
To run inference, modify `configs/inference_irradiance.yaml` to have the model configuration you want,
and `irradiance/model_inference.py` to run the current-best model on the data that you want.
Then run `python run.py --config-name inference_irradiance` to run inference. By default it uses the 4th GPU on Donatello.