Abstract: This specification provides a training method of a hybrid graph neural network model. The hybrid graph neural network model includes an encoding function and a decoding function. The method includes the following: using instances corresponding to all targets in training samples and several nearest neighbors of the instances as nodes in a graph, a graph representation vector of each instance is generated by using the encoding function based on graph data of all the instances. t rounds of training are performed on a decoding parameter; and in each round, bs targets are extracted from training samples, a predicted quantity of each target is generated by using the decoding function based on the graph representation vector of the instance corresponding to each target and non-graph data corresponding to each target, and the decoding parameter is optimized based on a loss quantity of the current round that is determined by the predicted quantities and label quantities of the bs targets in the current round.

Abstract: Embodiments of the present disclosure provide methods, devices and computer readable media for communication.

Abstract: Embodiments of the present disclosure disclose a method and apparatus for outputting information. A specific embodiment of the method comprises: connecting a database, in response to receiving a request for detecting the database; performing a write operation on a heartbeat table in the database, in response to detecting the connection with the database being normal; performing a read operation on the heartbeat table in the database, in response to detecting the write operation for the heartbeat table being abnormal; performing on the database an operation of writing a file to a hard disk, in response to detecting the read operation on the heartbeat table being normal; and outputting first information for representing a cause of a database failure, in response to detecting the operation of writing the file to the hard disk being abnormal. This embodiment improves the accuracy of the detection on the database failure.

Abstract: Embodiments of the present disclosure provide methods, devices and computer readable media for communication.

Abstract: Embodiments of the present disclosure disclose a method and apparatus for outputting information. A specific embodiment of the method comprises: connecting a database, in response to receiving a request for detecting the database; performing a write operation on a heartbeat table in the database, in response to detecting the connection with the database being normal; performing a read operation on the heartbeat table in the database, in response to detecting the write operation for the heartbeat table being abnormal; performing on the database an operation of writing a file to a hard disk, in response to detecting the read operation on the heartbeat table being normal; and outputting first information for representing a cause of a database failure, in response to detecting the operation of writing the file to the hard disk being abnormal. This embodiment improves the accuracy of the detection on the database failure.

Abstract: A substrate structure with invisible inner electrode patterns is provided. The substrate structure includes a substrate, an electrode pattern, a first laminated structure, and a passivation layer. The electrode pattern is disposed on the substrate. The first laminated structure is disposed on the electrode pattern, in which the first laminated structure includes a first upper layer, a second upper layer, and a third upper layer. The first upper layer is adjoined to the electrode pattern and the substrate. The first upper layer, the second upper layer, and the third upper layer are stacked sequentially. The passivation layer is disposed on the first laminated structure. The refractive indexes of the electrode pattern, the first upper layer, the second upper layer, the third upper layer, and the passivation layer are decreased sequentially.

Abstract: A substrate structure with invisible inner electrode patterns is provided. The substrate structure includes a substrate, an electrode pattern, a first laminated structure, and a passivation layer. The electrode pattern is disposed on the substrate. The first laminated structure is disposed on the electrode pattern, in which the first laminated structure includes a first upper layer, a second upper layer, and a third upper layer. The first upper layer is adjoined to the electrode pattern and the substrate. The first upper layer, the second upper layer, and the third upper layer are stacked sequentially. The passivation layer is disposed on the first laminated structure. The refractive indexes of the electrode pattern, the first upper layer, the second upper layer, the third upper layer, and the passivation layer are decreased sequentially.