Abstract: The present disclosure describes techniques for determining performance of a classifier. A first machine learning model and a second machine learning model may be trained by aggregating updates to the first machine learning model and the second machine learning model received from a plurality of client computing devices. A cumulative distribution function (CDF) associated with a distribution of the positive samples in the user data may be estimated using the trained first machine learning model. A probability density function (PDF) associated with a distribution of the negative samples in the user data may be estimated using the trained second machine learning model. An integration-based computation of an area under the receiver operating characteristic curve (AUC) of the classifier may be performed using the PDF and the CDF.

Abstract: The disclosure provides a method of treating a polyamine imbalance-related disorder. The method comprises administering phenylbutyrate to a subject in need thereof, thereby treating the polyamine imbalance-related disorder.

Abstract: A best available video stream is determined for each of multiple conference participants within a conference room including multiple cameras based on scores determined for video streams obtained from the cameras. The scores are determined based on representations of the conference participants within the video streams, for example, based on percentages of conference participant faces visible within the video streams, directions of conference participant faces relative to the cameras, directions of eye gaze of the conference participants relative to the cameras, and/or degrees to which conference participant faces are obscured within the video streams. The best available video streams are output for rendering within separate user interface tiles of conferencing software.

Abstract: In an embodiment a screen projection method includes determining, by a project end, to-be-projected data when a screen projection function is enabled, obtaining, by the project end, first data permission of the project end for the to-be-projected data and second data permission of a receive end for the to-be-projected data when the to-be-projected data is media data and projecting, by the project end, the to-be-projected data onto the receive end in a screen mirroring manner when the first data permission is higher than the second data permission or projecting, by the project end, the to-be-projected data onto the receive end in a digital living network alliance manner when the first data permission is lower than the second data permission.

Abstract: A device includes one Wi-Fi chip and a first antenna and a second antenna connected to the Wi-Fi chip. The first antenna wirelessly communicates with another device in a Wi-Fi P2P manner through a first channel. The second antenna wirelessly communicates with a first wireless signal source in a Wi-Fi AP manner through a second channel. After the first channel and the second channel are two different channels on a same frequency band, time division multiplexing is performed on the first antenna and the second antenna, and a second wireless signal source whose frequency band is different from that of the first wireless signal source is detected within preset duration, the second channel is adjusted to a channel which is different from the first channel on different frequency bands, and the second antenna wirelessly communicates with the second wireless signal source in the Wi-Fi AP manner through the adjusted second channel.

Abstract: A temperature prediction method and apparatus are provided. The method includes: determining a loss of a motor based on information about a motor controller, where the loss of the motor includes a first loss and a second loss, and the first loss is a loss generated by a fundamental wave component of a current of the motor (S210); and determining temperature of the motor based on the loss of the motor and a temperature prediction model (S220). According to the method, precision of temperature prediction of the motor can be improved.

Abstract: A method and system for building a thermal model of a power lithium-ion battery based on an electrochemical mechanism. The method includes: discretizing a second order partial differential heat conduction equation of a power lithium-ion battery according to a finite differential method, thereby building a thermal model of the power lithium-ion battery; carrying out a dynamic working condition test by using a cylindrical power lithium-ion battery selected as an object, thereby acquiring experimental data such as a temperature, a current, a voltage, and a temperature of a surface of the battery; identifying an electrochemical parameter of the power lithium-ion battery according to an optimal parameter algorithm by using test data acquired in a dynamic working condition, thereby building a thermal model of the power lithium-ion battery; and verifying accuracy of the thermal model of the power lithium-ion battery by using test data acquired in another dynamic working condition.

Abstract: A packet sending method includes, when a network element in a serial network receives a to-be-forwarded packet at a previous node from a previous network element, and when the network element needs to send a to-be-sent packet at a current node, determining whether the to-be-sent packet at the current node and the to-be-forwarded packet at the previous node can be superimposed, where the network element is an intermediate network element in the serial network; when the to-be-sent packet at the current node and the to-be-forwarded packet at the previous node can be superimposed, superimposing the to-be-sent packet at the current node and the to-be-forwarded packet at the previous node, to obtain a to-be-forwarded packet at the current node; and sending the to-be-forwarded packet at the current node to a next network element of the network element.

Abstract: A data transmission method and a device are provided. When determining that the first network element and the second network element is disconnected, the first network element stores identification information of the last frame of data and connection status information that are transmitted before the connection is disconnected. When determining a condition for re-communication is meet, the first network element generates a third key based on a first key and a second key, where the first key is a key used to transmit the last frame of data, and the second key includes a key used by the first network element and the second network element to transmit each of j frames of data before the last frame of data. The first network element communicates with the second network element based on the third key, identification information of an ith frame of data, and the connection status information.

Abstract: A method for selecting a management frame antenna based on a master-slave network and an apparatus are provided. In one example method, when a preset condition is met, a signal quality collection instruction is sent by a master device to a first slave device and a second slave device, so as to receive returned signal quality information. A management frame antenna is selected for the first slave device and a management frame antenna for the second slave device according to the received signal quality information, so that the selected management frame antennas can implement signal coverage optimization in an entire area managed by the master device.

Abstract: Methods and terminals for receiving and transmitting data information are provided. The method for receiving data information includes: determining whether information of a wireless power receiving module of a mobile terminal, obtained within a power transmission range by a wireless power transmitting module, satisfies a preset condition; and in response to the information of the wireless power receiving module satisfying the preset condition, performing a preset function of the mobile terminal. The preset function includes receiving and outputting a prompt message that is originally supposed to be outputted on the mobile terminal.

Abstract: The present disclosure provides an electronic device including a first part having a first display unit, and a second part coupled to the first part. The second part includes an input device located in a first region adjacent to a coupling region between the first part and the second part, and a second display unit located in a second region away from the coupling region between the first and second parts.

Abstract: Example sleeping and wake-up methods and apparatuses of a master-slave network, and a power saving system of a master-slave network, are described. One example method includes obtaining running statuses of slave devices on links from a link management list. A target device is determined. The target device is a slave device whose running status is idle. If running statuses of the target slave device to an end slave device on a link on which the target slave device is located are all idle, it is determined that a sublink from the target slave device to the end slave device on the link on which the target slave device is located is a target sublink. A power saving message is sent to at least one slave device on the target sublink, so that the slave device enters a sleep and listen state when receiving the power saving message.

Abstract: Embodiments of the present disclosure disclose a data processing method, apparatus, and system. The method provided in the embodiments of the present disclosure includes receiving a data processing request sent by a client and obtaining a routing table according to the data processing request. The routing table stores structure information of a relational database. The relational database includes a plurality of sub-databases. The sub-database includes a plurality of partitioned tables, and the partitioned table includes a plurality of shards. The method also includes determining a target sub-database according to the data processing request and the routing table. The method further includes sending the data processing request to the target sub-database, so that the target sub-database performs an operation according to the data processing request.

Abstract: The present invention is applicable to the field of communications technologies, and provides a repeating method of a wireless repeating device and a wireless repeating device. The method includes the following steps: synchronizing network connection information of an upstream AP to a downstream WLAN AP interface of the wireless repeating device; obtaining, according to the downstream WLAN AP interface corresponding to the synchronized upstream AP, downstream wireless STA MAC address information, and establishing an upstream WLAN Client interface which is in a mapping relationship with an STA MAC address; and performing, according to the mapping relationship between the STA MAC address and the established upstream WLAN Client interface, management on data forwarded between the upstream AP and a downstream wireless STA.

Abstract: The present invention provides an IP address allocation method for a master-slave network, an apparatus, and a system. A master device generates an IP address sub-segment group, and separately sends at least one IP address sub-segment in the IP address sub-segment group to at least one slave device, so that the slave device can use a received IP address sub-segment as a DHCP server address and allocate an IP address to a user device. In this way, the slave device can directly allocate the IP address to the user device, so that an IP address allocation time is reduced.

Abstract: A handover method in a master-slave network, and a master device, and slave devices are provided. The method includes when a master device or a first slave device determines that a user device needs to be handed over to a second slave device, the first slave device sends, to the second slave device, a first identifier of a first network interface allocated by the first slave device to the user device, and first network connection information used after the user device establishes a connection to the first network interface of the first slave device. The first identifier of the first network interface and a second identifier of the user device correspond on a one-to-one basis. The first slave device deactivates the first network interface and the second slave device activates a second network interface.

Abstract: A method for triggering a registrar protocol interaction. The method includes: receiving, by an access point, a probe request sent by at least one station, where the probe request includes an identifier of the access point; determining, by the access point, a first station in the at least one station according to the probe request sent by the at least one station, where the identifier included in the probe request sent by the first station matches an identifier of the access point, and a signal strength of the probe request sent by the first station is greater than a signal strength threshold; and automatically triggering, by the access point, a registrar protocol interaction with the first station.

Abstract: Example sleeping and wake-up methods and apparatuses of a master-slave network, and a power saving system of a master-slave network, are described. One example method includes obtaining running statuses of slave devices on links from a link management list. A target device is determined. The target device is a slave device whose running status is idle. If running statuses of the target slave device to an end slave device on a link on which the target slave device is located are all idle, it is determined that a sublink from the target slave device to the end slave device on the link on which the target slave device is located is a target sublink. A power saving message is sent to at least one slave device on the target sublink, so that the slave device enters a sleep and listen state when receiving the power saving message.