Forecasting load and charging requirements for vehicle fleets in companies.
You can find a detailed Documentation under docs/.
- Install Python
- Windows: https://www.python.org/
- MacOS:
brew install python - Linux:
- Debian/Ubuntu:
sudo apt install python3 - Fedora:
sudo dnf install python3 - Arch:
sudo pacman -S python
- Debian/Ubuntu:
- Clone the Repository
git clone https://github.com/DataDrivenSustainabilitySolutions/BiFlex-Forecasting.git - Install the Requirements
pip install -r requirements.txt
With Hyperparameter optimization:
python main.py \
--csv 'data/univariant_example.csv' \
--input_length 10 \
--prediction_length 1 \
--epochs 1 4 1 \
--batch_size 2 2 2 \
--test_size 0.2 \
--hyper_parameter_optimizer 'RandomSearch' \
--n_splits 2 \
--objective 'val_loss' \
--maxtrials 2 \
--should_optimize \
--metrics 'R2Score' \
--units 10 20 10 \
--activation 'relu' 'tanh' \
--dropout 0.0 0.5 0.1 \
--lamda 0.0005 0.001 0.0015 \
--model_optimizer 'Adam' \
--loss 'mse' \
--learning_rate 1e-4 1e-3 1e-2 \
--project_name 'project_complete_test_70trials'Without Hyperparameter optimization:
python main.py \
--csv 'data/univariant_example.csv' \
--input_length 10 \
--prediction_length 1 \
--epochs 1 \
--batch_size 2 \
--test_size 0.2 \
--hyper_parameter_optimizer 'RandomSearch' \
--n_splits 2 \
--objective 'val_loss' \
--maxtrials 2 \
--metrics 'R2Score' \
--units 10 \
--activation 'relu' \
--dropout 0.0 \
--lamda 0.0005 \
--model_optimizer 'Adam' \
--loss 'mse' \
--learning_rate 1e-4 \
--validation_split 0.2 \
--project_name 'project_complete_test_70trials'Contributions are what make the open source community a vibrant space for learning, inspiration, and creativity. Every contribution is deeply valued and appreciated.
For further Questions please contact one of them: