If you are here to see the predictions and the model features (with corresponding feature importance) you can find it in this notebook.
- You will need
python3
(preferablypython3.8
) to run this code. - If you use Ubuntu/Linux you can run the below commands in the terminal to get started.
- If you use another operating-system, you can use your preferred python IDE to set this project up with a virtual environment and install dependencies from the
requirements.txt
file.
Note regarding meteorological data used: The latest meteorological data is automatically extracted from the code. The weather forecast for March 1 - March 7, 2022 is prone to change and may slightly effect the predicted results. If you want to use the data we used, pass read_meteo_from_disk=True
when calling generate_prediction_file()
function.
If you use https
:
git clone https://github.com/ankitdhall/predicting-peak-bloom.git
cd predicting-peak-bloom
If you use ssh
:
git clone git@github.com:ankitdhall/predicting-peak-bloom.git
cd predicting-peak-bloom
If you want to download a .zip
:
Alternatively, you can download the zip from the github website.
pip3 install virtualenv
virtualenv my_venv
source my_venv/bin/activate
pip3 install -r requirements.txt
Run this script to make predictions for 2022-2031.
cd utils
python3 make_predictions.py
You will find the predictions in the predictions
folder as a predicted_bloom_doy.csv
file.
When you are in the parent directory run the following:
cd notebooks
jupyter-lab
Your browser should open with the jupyter-lab
interface.
- Click on the "Folder" icon on the top-left and select the "Predictions.ipynb".
- Once the notebook is open you can run the cells ("Kernel" > "Restart Kernel and run all cells...") to see the plots and predictions.
- The features and their importance to predict the DOY for each model.