Vertical K-means

Introduction

The calculation process for vertical Kmeans (two parties in this example) is illustracted as follows, where A and B are the trainer and the label_trainer who own the data, C is the assist_trainer who performs aggregation operations.

Parameters List

identity: str The role of each participant in federated learning, should be one of label_trainer, trainer or assist trainer.

model_info:

• name: str Model name, should be vertical_kmeans.

input:

• trainset:

  • type: str Train dataset type, currently supported is csv.

  • path: str The folder path of train dataset.

  • name: str The file name of train dataset.

  • has_id: bool Whether the dataset has id column.

  • has_label: bool Whether the dataset has label column.

output:

• path: str Output folder path.

• model:

  • name: str File name of output model.

• result:

  • name: str File name of result.

• summary:

  • name: str File name of summary.

train_info:

• train_params:

  • init: str Initialization method, should be one of random and kmeans++.

  • encryption:

    • otp:

      • key_bitlength: int Key length of one time pad encryption, support 64 and 128.

      • data_type: str Input data type, support torch.Tensor and numpy.ndarray, depending on model data type.

      • key_exchange:

        • key_bitlength: int Bit length of the key, recommend to be greater than or equal to 2048.

        • optimized: bool Whether to use optimized method.

      • csprng:

        • name: str Pseudo-random number generation method.

        • method: str Corresponding hash method.

  • k: int Number of clusters.

  • max_iter: int Maximum iteration.

  • tol: float Convergence threshold.

  • random_seed: int Random seed.