* @brief Split the dataframe along with the specified column * * @param df * @param partition * @param column * @return std::pair<std::vector<int

* @brief Splits the dataframe along with categorical column * * @param df * @param partition * @param column * @return std::pair<std::vector<int>

* @brief Split the dataframe along with continuous number column * * @param df * @param partition * @param column * @return std::pair<std::vector

Computes the total variance of an input tensor. Args: x (torch.Tensor): The input tensor. Returns: torch.Tensor: The total v

Train attacker models

Train clients on the public dataset. Returns: List[float]: a list of average loss of each client.

Train the global model on the public dataset. Returns: float: average loss

train shadow models Args: X (np.array): training data for shadow models y (np.array): training label for shadow model

Register the predicted logits to self.predicted_values

Update the aggregated consensus logits with the output logits received from the clients. Raises: NotImplementedError: Raises when

Updates the discriminator

Updates the Generator Args: batch_size (int): batch size epoch (int): epoch log_interval

Train the global model with the global consensus logits. Args: global_optimizer (torch.optim.Optimizer): an optimizer Re

Updates weight for each client given the received local gradients.

namespace storage

Node(){};