Distribute the current global model to each client. Args: force_send_model_state_dict (bool, optional): If True, send the global

Download the new global model

Calculate the Gaussian delta term. Args: delta (float): Delta value. Returns: float: Gaussian delta term.

Return epsilon of FedMD-NFDP Args: n (int): training set size k (int): sampling size replacement (bool, optional): sampli

Return k of FedMD-NFDP Args: epsilon (float): epsilon n (int): training set size replacement (bool, optional): sampling w

* @brief Get the number of rows * * @return size_t */

* @brief Inserts a categorical value * * @param column * @param value */

Returns the local performance of each clients. Returns: Dict[str, int]: performance of global model and local models

Update the global model with the local gradients.

Update the global model with the local model parameters.

Upload the local gradients

Calculate deviation of q Q = yy.T * K denotes the label - annotated version of K, and α denotes the SVM’s dual variables corresp

repeated .storage.TupleData data = 3;

Subtract two numbers in the log space. Answer must be non-negative.

Initializes the image and label, and set the optimizer Args: received_gradients: a list of gradients received from the client bat

Returns a closure function for the optimizer Args: optimizer (torch.optim.Optimizer): an instance of the optimizer fa

Aggregate the received logits with SA (calculating average)

Performs step 1 of the Hungarian algorithm. Subtracts the minimum value of each row from each element of the row, and then su

Assert the multi-box condition for a list of boxes. Args: boxes (list): A list of boxes. The format of each box is a tuple or list w

Calculate the Clopper-Pearson confidence interval. Args: num_success (int): Number of successes. num_total (int): Total numb

Computes the optimal assignment using the Hungarian algorithm. Args: mat (numpy.ndarray): The input matrix. Ret

Check for flickering between the current frame and previous frames. Args: frames (list): A list of frames. boxes_of_frames (

train an attacl model with the result of shadow models Args: shadow_result (dict): key is each class

Trains shadow models and get prediction of shadow models and membership label of each prediction for each class Args: