53results about How to "Guaranteed data volume" patented technology

The invention discloses an abnormal signal perception based abnormal step size acquisition method for vehicle electrical performance data. According to vehicle electrical performance data acquisitionrequirements, acquisition point pretreatment and an acquisition time prediction mechanism are added, and vehicle electrical performance data acquisition is converted into abnormal signal perception based abnormal step size acquisition; the pretreatment mechanism of acquisition points is defined, signal deviation estimation [rho]c and abnormal estimation which is shown in the description are introduced, and an abnormal signal in the acquisition points is corrected through the sampling of the acquisition points; and a signal margin which is shown in the description is defined, a signal bias coefficient a, zero drift b and a signal skipping coefficient which is shown in the description are introduced, and a real-time acquisition time prediction mechanism is designed. Through the pretreatmenton data, the acquisition time can be predicted to prepare the next acquisition, so that the vehicle electrical performance data abnormal step size acquisition can be realized; and the method can correct abnormal points during acquisition and dynamically change sampling step sizes, so that situation of high hardware requirements for high frequency, fixed cycle and longtime acquisition can be avoided, equipment costs can be reduced, and the efficiency and reliability of vehicle electrical performance data acquisition can be enhanced.

The invention relates to the field of business security of computer security, and discloses a risk model training method and device, computer equipment and a readable storage medium, and the method comprises the steps: obtaining long-term historical data through a long-term extraction rule, and obtaining recent historical data through a recent extraction rule; summarizing the long-term historical data and the recent historical data to obtain comprehensive training data; adjusting data volumes of long-term historical data and short-term historical data in the comprehensive training data, and adjusting a positive-negative proportion between a sum of a long-term positive sample and a short-term positive sample in the comprehensive training data and a sum of a long-term negative sample and a short-term negative sample in the comprehensive training data to obtain comprehensive adjustment data; and training the initial intelligent model through comprehensive adjustment data to obtain a mature model. The invention further relates to a block chain technology, and information can be stored in a block chain node. According to the invention, the risk identification efficiency is high, the accuracy is high, and the problems that experience and rules become vulnerabilities and normal users are identified as risk users are avoided.

The invention discloses a popup window processing method and device, computer equipment and a storage medium. The pop-up window processing method comprises the steps of monitoring a calling event of afunction module triggered by an online user, and obtaining triggering data of the function module; when the function module is a new function module, judging whether the trigger data meets a triggercondition of an initial popup event or not; popping up an evaluation popup window corresponding to the function module when the trigger data meets the trigger condition of the initial popup window event; when the triggering condition of the initial popup window event is not met, judging whether the triggering data meets the triggering condition of the popup window probability event or not; and when the triggering condition of the popup probability event is met, popping up the evaluation popup corresponding to the function module. The invention also relates to a blockchain technology, and the evaluation data can be stored in the blockchain. The method does not affect the use of the user on the basis of ensuring the data volume of the evaluation data.

The embodiment of the invention provides a model training method and device, computer equipment and a readable storage medium, and the method comprises the steps: obtaining an original sample set, training a sample recognition model according to the original sample set, and obtaining an original recognition model, wherein the original sample set comprises a plurality of original samples; calling the original identification model to identify each original sample to obtain an original identification result of each original sample; dividing the original sample set into a target sample set and a first activation sample set according to the original identification result of each original sample; calling an activation model to activate each target sample to obtain a second activation sample set;and training a recognition processing model according to the first activation sample set and the second activation sample set to obtain a target recognition model, so that the translation accuracy ofthe trained translation model is effectively improved while the data volume of the trained translation model can be ensured.

The invention discloses an image processing method and device, equipment and a medium, and the method comprises the steps: obtaining sample data, calling an image processing model to convert an annotated real face image Ai into a reconstructed cartoon face image B'i, and converting a non-annotated real face image Ci into a reconstructed cartoon face image B''i; inversely converting the reconstructed cartoon face image B''i into a reconstructed real face image C''i; inversely converting the marked cartoon face image Bi into a reconstructed real face image C'i; obtaining a first loss according to the difference between B'i and Bi; obtaining a first loss according to B'i, B''i and Bi, obtaining a second loss according to B'i, B''i and Bi, obtaining a third loss according to Ci and C'i, obtaining a fourth loss according to the difference between Ci and C''i; and training an image processing model based on the first loss, the second loss, the third loss and the fourth loss. By means of the method, the model training effect can be improved, and the universality and stability of the trained image processing model are improved.

An embodiment of the invention provides a flight equipment communication method and equipment, a server and a storage medium. The method comprises the steps that: target flight equipment sends an access request message to a plurality of adjacent nodes, and receives channel state information returned by the plurality of adjacent nodes based on the access request message, wherein the adjacent nodes comprise other flight equipment adjacent to the target flight equipment and/or a base station; according to the channel state information, link communication information between the target flight equipment and each adjacent node is determined; according to the link communication information, at least one candidate access node is determined from the plurality of adjacent nodes, and the initial transmitting power and the initial data transmission rate of the target flight equipment corresponding to each candidate access node are respectively determined; and a target access node is determined from the at least one candidate access node based on the initial transmitting power and the initial data transmission rate, and the target flight equipment is controlled to access the target access node. Therefore, the data transmission quality and data volume of the flight equipment are ensured.