Abstract: A method for scaling the porcelain material of a restoration and the associated machining process is invented, which comprises: establishing a data model of the restoration (S100); dividing the bonding surface or interface between veneer porcelain to be prepared and a restoration substrate into a finite number of element surfaces (S200); establishing the direction coordinate axis on every element surface as its reference axis for setting up a size and a thickness value of veneer porcelain of each element surface (S300); determining and specifying the size and thickness of the veneer porcelain of every element surface according to the preset size and thickness value, and completing the bonding surface by covering it with the veneer porcelain composed of all the element surfaces as scaled with those sizes and thickness values as referred to (S400); generating the outer contour of the veneer porcelain of a clinic restoration (S500).

Abstract: A federated learning training method is performed by a server. The method comprises: receiving gradient information of a labeled sample of each client sent by each client; according to the gradient information sent by each client, obtaining target gradient information belonging to a same labeled sample; determining a client to which each labeled sample belongs; and sending, to a client to which the same labeled sample belongs, the target gradient information corresponding to the same labeled sample.

Abstract: In a method for training a federated learning model, a server obtains a target split mode corresponding to a training node in response to determining that the training node satisfies a preset splitting condition. The server notifies a client to perform, based on the target split mode, node splitting. The server performs a next round of training by taking a left subtree node generated by performing the node splitting as a new training node until an updated training node does not satisfy the preset splitting condition. The server performs a next round of training by taking another non-leaf node of the boosting tree as a new training node. The server stops training and generates a target federated learning model in response to determining that a node dataset of the plurality of boosting trees is empty.

Abstract: A method and a device for testing a memory, and a non-transitory readable storage medium are provided. The method for testing the memory includes: executing a memory testing program to perform a memory test on a first area of the memory, the first area being an area that is not occupied by the memory testing program; writing address information of a second area into an external memory of a device, the second area being an area that is occupied by the memory testing program; transferring, after the memory test on the first area is completed, the memory testing program into a partial area of the first area according to the address information of the second area recorded in the external memory; and executing the memory testing program to perform the memory test on the second area.

Abstract: A post package repairing method for a memory includes: during a memory test, writing failure information of the memory into the memory, the failure information including failure addresses; after a device is powered on, reading the failure addresses from the memory, and determining a number of failed lines where the failure addresses are located; when the number of the failed lines is less than or equal to a number of redundant lines, repairing the failed lines by using the redundant lines; and when the number of the failed lines is greater than the number of the redundant lines, loading the failure information into a register.

Abstract: A method for scaling the porcelain material of a restoration and the associated machining process is invented, which comprises: establishing a data model of the restoration (S100); dividing the bonding surface or interface between veneer porcelain to be prepared and a restoration substrate into a finite number of element surfaces (S200); establishing the direction coordinate axis on every element surface as its reference axis for setting up a size and a thickness value of veneer porcelain of each element surface (S300); determining and specifying the size and thickness of the veneer porcelain of every element surface according to the preset size and thickness value, and completing the bonding surface by covering it with the veneer porcelain composed of all the element surfaces as scaled with those sizes and thickness values as referred to (S400); generating the outer contour of the veneer porcelain of a clinic restoration (S500).