We present MC-ANN, a Mixture Clustering-Based Attention Neural Network for time Series Forecasting, which has been accepted by TPAMI 2025. If you make use of our code or data, please cite our paper.

@ARTICLE{li2025mcann,
  author={Yanhong Li and David Anastasiu}, 
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, 
  title={ MC-ANN: A Mixture Clustering-Based Attention Neural Network for Time Series Forecasting}, 
  year={2025}, 
  volume={},  
  number={},  
  pages={}, 
  keywords={ Pattern recognition; deep recurrent neural networks; encoder decoder; time series forecasting; attention clustering network; Gaussian Mixture; hydrology prediction; reservoir water level},
  doi={10.1109/TPAMI.2025.3565224}}

Experiments were executed in an Anaconda 3 environment with Python 3.8.8. The following will create an Anaconda environment and install the requisite packages for the project.

conda create -n MCANN python=3.8.8
conda activate MCANN
conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 -c pytorch
python -m pip install -r requirements.txt

Download the datasets from here and upzip the files in the data_provider directory. In the ./data_provider/datasets directory, there should now be 5 reservoir sensor (file names end with _sof24.tsv) datasets.

--reservoir_sensor: reservoir dataset file name. The file should be csv file.

--rain_sensor: rain dataset file name. The file should be csv file.

--train_volume: train set size.

--hidden_dim: hidden dim of lstm layers.

--atten_dim: to set hidden dim of attention layers.

--layer: number of layers.

--os_s: oversampling steps.

--os_v: oversampling frequency.

--seq_weight: sequence cluster weight.

--watershed: 1 if trained with rain info, else 0.

--model: model name, used to generate the pt file and predicted file names.

--mode: set it to 'train' or 'inference' with an existing pt_file.

--pt_file: if set, the model will be loaded from this pt file, otherwise check the file according to the assigned parameters.

--save: if save the predicted file of testset, set to 1, else 0.

--outf: default value is './output', the model will be saved in the train folder in this directory.

Refer to the annotations in run.py for other parameter settings. Default parameters for reproducing are set in the files (file names start with opt and end with .txt) under './models/'.

The Jupyter notebook example.ipynb shows how to train a model via command line commands and use specific model functions to perform inference on the Stevens Creek sensor dataset.