Abstract: Provided are systems for tuning prediction results of a machine learning model that include at least one processor to determine a plurality of values associated with a prediction matrix based on an output of a trained machine learning model, tune a set of reference measures to provide an adjustment to a predicted classification value of a prospective output of the trained machine learning model, apply the set of reference measures to determine a predicted classification value of a real-time output of the trained machine learning model, wherein the output of the trained machine learning model comprises a predicted classification value for a real-time event. Methods and computer program products are also provided.

Abstract: Methods, systems, and computer program products may obtain a machine learning model, a training dataset including a time range and a feature set including a number of features, and a number times to split the training dataset; for each feature in the feature set, determine, based on a difference between a first trained model including the machine learning model trained on the training dataset with that feature and a second trained model including the machine learning model trained on the training dataset without that feature, whether to update the training dataset to include an updated feature set by removing the feature from the feature set of the training dataset; train the machine learning model on the training dataset including the updated feature set to generate a trained machine learning model; and provide the trained machine learning model.

Abstract: Embodiments of the present disclosure provide an earphone debugging method and device, and a storage medium. The earphone debugging method comprises the following steps: obtaining configuration parameters to be updated of a slave earphone and configuration parameters to be updated of a master earphone by the master earphone; sending, by the master earphone, the configuration parameters to be updated of the slave earphone to the slave earphone, which allows the slave earphone to be operated according to the configuration parameters to be updated of the slave earphone; and the master earphone being operated according to the configuration parameters to be updated of the master earphone. According to the embodiments of the present disclosure, the earphone debugging efficiency is improved.

Abstract: The present disclosure provides a content display method, a content display apparatus, an electronic device, a storage medium, and a program product. The method includes: receiving a display operation for a target multimedia content, the target multimedia content being a content associated with a preset keyword; and in response to the display operation, displaying the target multimedia content in a target content display region of a display page, and presenting an associated effect for the preset keyword in a first effect display region associated with the target content display region, where at least two content display regions are provided in a main display region of the display page, the at least two content display regions include the target content display region, and a content display region other than the target content display region is located outside the first effect display region.

Abstract: A luminous garment includes a garment body, a logo element, a power source, a control module, and an operation interface. The garment body is wearable by a user. The logo element is arranged on the surface of the garment body to emit light upon receiving an electrical current flowing therethrough. The operation interface is arranged for operation by the user to generate an operation signal and transmit the operation signal to the control module. The control module is arranged to control, according to the operation signal, a light emission mode of the logo element.

Abstract: An information platform for providing an earthquake early warning of a large area is disclosed. The large area includes a plurality of small districts, different sets of the plurality of small districts form different large districts according to geographical locations so as to make up the large area, and the information platform is both connected with a plurality of on-site earthquake early warning stations and the plurality of small districts through a cloud network. The plurality of on-site earthquake early warning stations are located in the large districts respectively, each of the plurality of on-site earthquake early warning stations is configured to obtain a real-time seismic longitudinal wave measurement data, obtain a real-time seismic transverse wave feature prediction value corresponding to the real-time seismic longitudinal wave measurement data, and transmit the real-time seismic transverse wave feature prediction value.

Abstract: A method of fabricating a metal mask includes receiving a metal planar substrate and patterning the metal planar substrate. The metal planar substrate includes a first surface and a second surface opposite to the first surface. The patterning the metal planar substrate includes forming strip-shaped structures, forming through holes, and forming a blind hole in a direction from the first surface to the second surface. The through holes extend to the first surface and the second surface. The through holes and the strip-shaped structures are alternately arranged. The blind hole extends across the through holes.

Abstract: The present invention belongs to the technical field of preparation of light metal alloy materials, in particular to a method for producing a magnesium-lithium alloy by a gaseous co-condensation method. The method comprises the steps of: 1) mixing and briquetting a lithium salt, a refractory agent and a catalyst under pressure, and then thermally decomposing to form an unsaturated composite oxide; 2) respectively crushing and ball-milling, and then briquetting the unsaturated composite oxide, magnesium oxide, a reducing agent and a fluxing agent; 3) reducing briquettes in vacuum; 4) making a gas pass through a first condensing chamber of a temperature control device, and then purifying; 5) The purified metal gas is condensed into the condensing phase of the alloy through the second condensing chamber of a quenching device; 6) obtaining the magnesium-lithium alloy with a purity being 99.5% or above by virtue of smelting and flux-refining, and then purifying by distillation.

Abstract: This application provides a communication method and apparatus. The method includes: receiving first information; determining N modes from M modes based on the first information, where N is an integer greater than 0, M is an integer greater than or equal to N, and the M modes include one or more of a first mode, a second mode, and a third mode; and sending a transport block in one of the N modes. A transmission mode of at least one antenna panel on a terminal device is accurately indicated by indication information.

Abstract: This application provides a communication method, apparatus, and system, so that each antenna panel can perform independent power control, making power control more accurate, and improving information transmission efficiency. The method includes: A terminal device obtains a first parameter corresponding to each of a plurality of antenna panels; determines, based on the first parameter corresponding to each of the plurality of antenna panels, a power control parameter set corresponding to each of the plurality of antenna panels; determines, based on the power control parameter set corresponding to each antenna panel, power for transmitting information on each antenna panel; and transmits information on each antenna panel based on the power for transmitting information on each antenna panel.

Abstract: A method for communicating between a wireless earphone and a charging case, a charging case, a TWS earphone and a TWS earphone system are provided. In the method, in response to a case opening signal, the earphone case I/O controlling module is controlled to set an operation attribute of the first I/O interface as a communication interface. A predetermined negotiation signal is transmitted to the wireless earphone via the first I/O interface. It is determined whether a negotiation confirmation signal corresponding to the predetermined negotiation signal is received from the wireless earphone via the first I/O interface. In a case that the negotiation confirmation signal corresponding to the predetermined negotiation signal is received from the wireless earphone via the first I/O interface, the charging case enters into a communication mode for communicating with the wireless earphone.

Abstract: A request to configure a connect cluster including one or more connectors for a cloud computing system is received. The request include a desired connector state. A connector specification file is automatically generated based on the desired connector state for the connect cluster via a declarative application programming interface (API). Application resources associated with the connect cluster are automatically configured based on the specification file.

Abstract: Provided is a system for segmenting large scale datasets according to machine learning models based on transfer learning that includes at least one processor programmed or configured to train a base machine learning model using a training dataset to generate a trained machine learning model, evaluate the trained machine learning model using an evaluation dataset, wherein, when evaluating the trained machine learning model using the evaluation dataset, the at least one processor is programmed or configured to generate a confidence score for each data instance of the evaluation dataset with the trained machine learning model, augment the evaluation dataset based on the confidence score for each data instance of the evaluation dataset to generate an augmented evaluation dataset, and retrain the trained machine learning model using the augmented evaluation dataset to generate a final machine learning model. Methods and computer program products are also provided.

Abstract: A deployment request is received for a plurality of clusters within a plurality cloud computing environments. The deployment request relates to one or more operational actions for computing devices associated with the plurality of clusters. The plurality of clusters is associated with a single application programming interface (API). A set of standards are received for the plurality of clusters from the API. The plurality of clusters is configured according to the set of standards to create a configured plurality of clusters. The configured plurality of clusters are deployed within the plurality of cloud computing environments.

Abstract: An earthquake detection method is disclosed. The method comprises the following steps: receiving a first signal and at least a second signal, wherein the first signal is triggered in response to a first vibration state of a main detection site, and the at least a second signal is triggered in response to at least a second vibration state of at least a backup detection site; determining whether a quantity of those being enabled among the at least a second signal satisfies a quantity ratio condition, and determining whether the first signal and the quantity of the at least a second signal are both enabled at a specific time point, wherein the first and the quantity of the at least a second signals respectively have a first and at least a second initial enabled time points; and determining there is a real earthquake event when an interval between the first and the at least a second signal satisfies a time relationship.

Abstract: In the present invention, an earthquake detection system based on the combination configuration of a free field and remote signal source is provided. The earthquake detection system includes: a host; a main sensor arranged in a free field and connected to the host; and a plurality of backup sensors, which are remotely arranged with respect to the main sensor, wherein each of the backup is connected to the host via the internet and transmits a remote signal to the host so as to avoid misjudging the occurrence of earthquakes because of unnatural factors, and prevent the vibration caused by human activities from interfering with the detector. The earthquake detection system achieves the effect of verification through a plurality of sensors installed in different positions.

Abstract: An earthquake detection system is disclosed. The earthquake detection system includes a main sensor, a first backup sensor, a second backup sensor and a third backup sensor. The main sensor is disposed on a free field. The first backup sensor is disposed on a structure. The second backup sensor is disposed below a ground surface at a depth of at least 20 meters. The third backup sensor is disposed at a remote end relative to the main sensor, and outputs a remote signal through a network so as to avoid misjudging the occurrence of earthquakes because of unnatural factors, and prevent the vibration caused by human activities from interfering with the detector. The earthquake detection system achieves the effect of verification through a plurality of sensors installed in different positions. Only when all the sensors confirm that there is an earthquake, an earthquake warning will be issued to the protected place.