Time Series Forecasting
Transformers
Safetensors
fela_grid_renewable
feature-extraction
fela
fourier-neural-operator
fno
cpu
on-device
energy-forecasting
solar-power
wind-power
probabilistic-forecasting
quantile-regression
custom_code
Instructions to use lowdown-labs/fela-power-grid with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use lowdown-labs/fela-power-grid with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("lowdown-labs/fela-power-grid", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "model_type": "fela_grid_renewable", | |
| "description": "Dual path FNO probabilistic power forecaster for solar and wind, trained on GEFCom2014. Outputs 99 quantiles per hour from weather covariates.", | |
| "architecture": "fno_dual_path", | |
| "framework": "pytorch", | |
| "architectures": [ | |
| "FelaGridModel" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "configuration_grid.FelaGridConfig", | |
| "AutoModel": "modeling_grid.FelaGridModel" | |
| }, | |
| "default_track": "solar", | |
| "tracks": { | |
| "solar": { | |
| "params": 136780, | |
| "input_steps": 6, | |
| "input_features": 20, | |
| "input_shape": [1, 6, 20], | |
| "weights_safetensors": "solar.safetensors", | |
| "dims": {"Fin": 20, "L": 6, "D": 64, "modes": 3, "nblk": 4, "nq": 99, "arch": "dual"} | |
| }, | |
| "wind": { | |
| "params": 311554, | |
| "input_steps": 12, | |
| "input_features": 15, | |
| "input_shape": [1, 12, 15], | |
| "weights_safetensors": "wind.safetensors", | |
| "dims": {"Fin": 15, "L": 12, "D": 96, "modes": 6, "nblk": 3, "nq": 99, "arch": "dual"} | |
| } | |
| }, | |
| "num_quantiles": 99, | |
| "quantile_levels_percent": "1..99", | |
| "output_shape": "[batch, 99]", | |
| "output_units": "power as fraction of site rated capacity (0 to 1), 99 monotone quantiles for the center forecast hour", | |
| "preprocessing": "standardize the NWP window per feature; RevIN runs inside the model. See modeling.preprocess_nwp", | |
| "license": "lowdown-labs-lovely-license-1.0" | |
| } | |