Abstract: A method, an apparatus, and a system for generating a neural network model, a device, a medium, and a program product are provided. In an embodiment, a first device sends an indication about a structure of a subnetwork model to a second device, where the subnetwork model is determined by adjusting a structure of a hypernetwork model. The first device receives a parameter of the subnetwork model from the second device, where the parameter of the subnetwork model is determined by the second device based on the indication and the hypernetwork model. The first device trains the subnetwork model based on the received parameter of the subnetwork model. The first device sends a parameter of the trained subnetwork model to the second device for the second device to update the hypernetwork model. In the foregoing manner, an efficient federated learning scheme between a plurality of devices is provided.

Abstract: A method for route handling of a regional centralized control station is based on a CTC3.0 technology, and includes the following steps: step (1): generating multi-station train and shunting plans according to a multi-station planning terminal, and executing steps (2) and (4) at the same time; step (2): sending the train plan to an autonomous computer to generate a train route sequence, and executing step (3); step (3): sending, by the autonomous computer, the train route sequence to a regional centralized control route handling terminal; step (4): compiling, by the multi-station planning terminal, an automatically generated shunting route sequence in the shunting plan, and executing step (5); and step (5): synchronizing, by a center service, the shunting route sequence to the regional centralized control route handling terminal.

Abstract: A flow splitting device for gas reverse circulation drilling, including: an upper joint for connecting with a double-wall drill pipe; an inner tube which is arranged in the upper joint and defines a first passageway in communication with an inner chamber of the double-wall drill pipe, a second passageway in communication with an annular space in the double-wall drill pipe formed between the inner tube and the upper joint; a lower joint, having an upper end fixedly connected with the upper joint and a lower end for connecting with a drill tool; and a flow guiding member provided between the upper joint and the lower joint. A flexible sealing mechanism is provided outside the upper joint. The flexible sealing mechanism extends radially outward relative to the upper joint and the lower joint to form a sealing contact with a wellbore wall.

Abstract: A system and method for implementing slot-relation extraction for a task-oriented dialogue system that includes implementing dialogue intent classification machine learning models that predict a category of dialogue of a single utterance based on an input of utterance data relating to the single utterance, wherein the category of dialogue informs a selection of slot-filling machine learning models; implementing the slot-filling machine learning models that predict slot classification labels for each of a plurality of slots within the utterance based on the input of the utterance data; implementing a slot relation extraction machine learning model that predicts semantic relationship classifications between two or more distinct slots of tokens of the utterance; and generating a response to the single utterance or performing actions in response to the single utterance based on the semantic relationship classifications between the distinct pairings of the two or more distinct slots of the single utterance.

Abstract: A system and method for implementing slot-relation extraction for a task-oriented dialogue system that includes implementing dialogue intent classification machine learning models that predict a category of dialogue of a single utterance based on an input of utterance data relating to the single utterance, wherein the category of dialogue informs a selection of slot-filling machine learning models; implementing the slot-filling machine learning models that predict slot classification labels for each of a plurality of slots within the utterance based on the input of the utterance data; implementing a slot relation extraction machine learning model that predicts semantic relationship classifications between two or more distinct slots of tokens of the utterance; and generating a response to the single utterance or performing actions in response to the single utterance based on the semantic relationship classifications between the distinct pairings of the two or more distinct slots of the single utterance.

Abstract: A system and method for implementing slot-relation extraction for a task-oriented dialogue system that includes implementing dialogue intent classification machine learning models that predict a category of dialogue of a single utterance based on an input of utterance data relating to the single utterance, wherein the category of dialogue informs a selection of slot-filling machine learning models; implementing the slot-filling machine learning models that predict slot classification labels for each of a plurality of slots within the utterance based on the input of the utterance data; implementing a slot relation extraction machine learning model that predicts semantic relationship classifications between two or more distinct slots of tokens of the utterance; and generating a response to the single utterance or performing actions in response to the single utterance based on the semantic relationship classifications between the distinct pairings of the two or more distinct slots of the single utterance.

Abstract: A system and method for implementing slot-relation extraction for a task-oriented dialogue system that includes implementing dialogue intent classification machine learning models that predict a category of dialogue of a single utterance based on an input of utterance data relating to the single utterance, wherein the category of dialogue informs a selection of slot-filling machine learning models; implementing the slot-filling machine learning models that predict slot classification labels for each of a plurality of slots within the utterance based on the input of the utterance data; implementing a slot relation extraction machine learning model that predicts semantic relationship classifications between two or more distinct slots of tokens of the utterance; and generating a response to the single utterance or performing actions in response to the single utterance based on the semantic relationship classifications between the distinct pairings of the two or more distinct slots of the single utterance.

Abstract: The invention belongs to the technical field of preparation of a ceramsite catalyst and discloses a ceramsite catalyst containing metal oxide particles and a preparation method and application thereof. The preparation method including the following specific steps of: dissolving a metal salt with an ozone catalytic activity in water to obtain a metal salt solution; mixing the metal salt solution with a ceramsite raw material to prepare a ceramsite embryoid body; and sintering the ceramsite embryoid body at a high temperature to obtain the ceramsite catalyst containing metal oxide particles. The ceramsite catalyst containing metal oxide particles of the invention can be repeatedly used, so that secondary pollution is reduced; and with the same ozone input quantity, by adding the ceramsite catalyst containing metal oxide particles of the present invention, the efficiency of water treatment through ozone oxidization can be increased by 41.35%.

Abstract: A software deployment method creates and provides an installation parameter file for each computer to be deployed according to a software deployment task. The installation parameter file of each computer to be deployed is used to guide a network installation of software on the computer. A corresponding software deployment system, a software deployment server, and a software deployment user server are also provided. The installation parameter files are generated collectively according to the software deployment tasks, so that the computer equipment is guided by the installation parameter file to install the software automatically. Therefore, the software deployment on the computers in batches is more convenient. Moreover, as the installation parameter file corresponds to each computer to be deployed, the software type and the parameter configuration of each computer to be deployed can be adjusted flexibly, which facilitates the customization of the software.

Abstract: A software deployment method creates and provides an installation parameter file for each computer to be deployed according to a software deployment task. The installation parameter file of each computer to be deployed is used to guide a network installation of software on the computer. A corresponding software deployment system, a software deployment server, and a software deployment user server are also provided. The installation parameter files are generated collectively according to the software deployment tasks, so that the computer equipment is guided by the installation parameter file to install the software automatically. Therefore, the software deployment on the computers in batches is more convenient. Moreover, as the installation parameter file corresponds to each computer to be deployed, the software type and the parameter configuration of each computer to be deployed can be adjusted flexibly, which facilitates the customization of the software.