Abstract: Methods, systems, electronic devices, and computer-readable storage media for image normalization processing are provided. In one aspect, an image normalization processing method includes: normalizing a feature map by respectively using K normalization factors to obtain K candidate normalized feature maps; for each of the K normalization factors, determining a first weight value for the normalization factor; and determining a target normalized feature map corresponding to the feature map based on the candidate normalized feature map corresponding to each of the K normalization factors and the first weight value for each of the K normalization factors. The K candidate normalized feature maps and the K normalization factors have a one-to-one correspondence, and K is an integer greater than 1.

Abstract: The present disclosure relates to a data processing method and apparatus, and a storage medium. The method includes: inputting input data into a neural network model to obtain feature data currently output by a network layer in the neural network model (S100); determining, according to transformation parameters of the neural network model, a normalization mode matched with the feature data (S200), wherein the transformation parameters are used for adjusting a statistical range of statistics of the feature data, and the statistical range is used for representing the normalization mode; and performing normalization processing on the feature data according to the determined normalization mode to obtain normalized feature data (S300). According to embodiments of the present disclosure, the purpose of autonomously learning a matched normalization mode for each normalization layer of the neural network model can be implemented without human intervention.

Abstract: Embodiments of the present disclosure disclose normalization methods and apparatuses for a deep neural network, devices, and storage media. The method includes: inputting an input data set into a deep neural network, the input data set including at least one piece of input data; normalizing a feature map set output by means of a network layer in the deep neural network from at least one dimension to obtain at least one dimension variance and at least one dimension mean; and determining a normalized target feature map set based on the at least one dimension variance and the at least one dimension mean. Based on the embodiments of the present disclosure, normalization is performed along at least one dimension so that statistics information of each dimension of a normalization operation is covered, thereby ensuring good robustness of statistics in each dimension without excessively depending on the batch size.

Abstract: The present invention discloses an automatic loading system and method for service functions of a hydraulic machine. The loading system includes an acquiring portion for acquiring i service function component chains of a hydraulic machine, a determining portion for determining a demand sequence of the service function component chains used by a user within each preset time period in a measuring cycle based on the acquired service function component chains and a deciding portion for automatically loading a service function of the hydraulic machine in a next moment according to the demand sequence based on a current input operation command of the user. By adopting the automatic loading system or method provided by the present invention, response speed of intelligent services of the hydraulic machine can be improved.

Abstract: The present invention discloses a method and system for determining a maintenance policy of a complex forming device. The method and system include: establishing a performance judgement standard; obtaining actual operation data; and determining a performance stage of the complex forming device, and determining the maintenance policy of the complex forming device. The establishing a performance judgement standard includes allocating data located at different performance stages to corresponding working condition units; setting data located at a normal performance stage to normal reference data, and calculating, by using a shortest path algorithm, distances between operation data of different working condition units in different functional modules and the normal reference data, to obtain performance statuses of the different functional modules.

Abstract: The present invention discloses a method and system for determining a maintenance policy of a complex forming device. The method and system include: establishing a performance judgement standard; obtaining actual operation data; and determining a performance stage of the complex forming device, and determining the maintenance policy of the complex forming device. The establishing a performance judgement standard includes allocating data located at different performance stages to corresponding working condition units; setting data located at a normal performance stage to normal reference data, and calculating, by using a shortest path algorithm, distances between operation data of different working condition units in different functional modules and the normal reference data, to obtain performance statuses of the different functional modules.

Abstract: The present invention discloses an automatic loading system and method for service functions of a hydraulic machine. The loading system includes an acquiring portion for acquiring i service function component chains of a hydraulic machine, a determining portion for determining a demand sequence of the service function component chains used by a user within each preset time period in a measuring cycle based on the acquired service function component chains and a deciding portion for automatically loading a service function of the hydraulic machine in a next moment according to the demand sequence based on a current input operation command of the user. By adopting the automatic loading system or method provided by the present invention, response speed of intelligent services of the hydraulic machine can be improved.

Abstract: Disclosed is an apparatus for troubleshooting fault component in equipment and a method thereof. The device includes portions for building a fault-component-sensor Bayesian belief network model by acquiring component state-abnormal and data of fault maintenance of the equipment, and for calculating probabilities of actual component abnormality when a sensor connected with an component detects component abnormality based on the model; and arranging the probabilities in a descending order to obtain the arranged probabilities of actual component abnormality, and the top-arranged component being the one to be troubleshot first. Namely, a relational expression among the fault, the component and the sensor may be systematically built by adopting the method or the apparatus provided by the present disclosure, and the most-likely-failing component in an equipment may be quickly detected according to the expression, thereby improving troubleshooting efficiency.

Abstract: The present invention relates to a method for identifying a preferred region for a product and an apparatus and a storage medium thereof. The method is executed by computer. The method includes: obtaining comment texts of users in different regions for a to-be-analyzed product, and extracting product features of the to-be-analyzed product from the comment texts; determining sentiment polarities of the users for the product features in the comment texts; calculating associations between sentiment orientations of the product features and the regions; extracting product features with regional preferences from the product features; and determining, for each product feature with a regional preference, a preferred region for the product feature in view of the sentiment polarities. The present invention can provide preferred regions for on-line product comments, enable enterprises to formulate more specific marketing strategies, and drive enterprises to implement regional product marketing strategies.

Abstract: The present invention relates to a method and device for on-line prediction of remaining driving mileage of an electric vehicle. The method comprises: acquiring in-transit data and driving environment data of the electric vehicle which is driving; calculating the power consumption per mileage of the electric vehicle in the current case by using the in-transit data and the driving environment data in combination with a power consumption rate data model; predicting the remaining driving mileage of the electric vehicle based on the power consumption per mileage. The device provided by the present invention is implemented on the basis of the method above. The prediction result of the present invention is more accurate, to avoid the problem that the power is exhausted due to exceeding the mileage expected by a user so that the electric vehicle cannot continue to drive, thereby improving the driving experience of the user.

Abstract: The present invention relates to a method and an apparatus for identifying a preferential region for a product. The method includes: obtaining comment texts of users in different regions for a to-be-analyzed product, and extracting product features of the to-be-analyzed product from the obtained comment texts; determining sentiment polarities of the users for the product features in the comment texts; calculating associations between sentiment orientations of the product features and the regions; extracting product features with regional preferences from the product features; and determining, for each extracted product feature with a regional preference, a preferential region for the product feature in view of the sentiment polarities. For content of fragmental and random online comments on the product, the present invention can provide a preferential region, enable an enterprise to formulate a more specific marketing strategy, and drive the enterprise to implement the regional product marketing strategy.

Abstract: The present invention relates to a method and device for on-line prediction of remaining driving mileage of an electric vehicle. The method comprises: acquiring in-transit data and driving environment data of the electric vehicle which is driving; calculating the power consumption per mileage of the electric vehicle in the current case by using the in-transit data and the driving environment data in combination with a power consumption rate data model; predicting the remaining driving mileage of the electric vehicle based on the power consumption per mileage. The device provided by the present invention is implemented on the basis of the method above. The prediction result of the present invention is more accurate, to avoid the problem that the power is exhausted due to exceeding the mileage expected by a user so that the electric vehicle cannot continue to drive, thereby improving the driving experience of the user.