34results about How to "Flexible training" patented technology

Embodiments of the invention provide a task scheduling method and apparatus, an electronic device and a computer readable storage medium. The method comprises the steps of obtaining a target task queue, wherein the target task queue is a task queue comprising at least one to-be-scheduled task; and sequentially scheduling the to-be-scheduled tasks in the target task queue according to the sequenceof the priorities of the to-be-scheduled tasks in the target task queue from high to low. According to the embodiment of the invention, task scheduling of different priorities can be supported, so that flexible task priority scheduling can be realized, and the flexibility of task scheduling is improved.

The invention relates to the field of deep learning image generation, in particular to a multi-domain image conversion method and system based on a conditional generative adversarial network, and a medium, and the method comprises the implementation steps: inputting to-be-converted x-modal original images x and y-modal original images y; adopting a pre-trained condition extractor to generate an xmodal condition Cx and a y modal condition Cy; and inputting the original image x, the original image y, the x modal condition Cx and the x modal condition y into a pre-trained condition generative adversarial network to obtain a corresponding image conversion result. According to the method, the characteristics of the original image are extracted by utilizing the characteristic extractor, the condition matrix is obtained through up-sampling and splicing with the zero matrix on the channel, and the semantic information of each modal input is kept under the condition of higher independence; andtraining is flexible, the number of domains to be converted is not limited, and few parameters are needed.

The invention discloses a method, a system and a medium for generating registered multi-modal MRI with focus segmentation tags. The method comprises the following steps: acquiring a random matrix in normal distribution, inputting the random matrix into a trained structural feature decoder in a generative adversarial network, and decoding to generate a structural feature map; fusing the structuralfeature map and the randomly selected focus segmentation label map through random input to obtain a fusion result, and inputting the fusion result into a trained random encoder in the generative adversarial network to obtain a code; and inputting the code into a decoder of each mode trained in the generative adversarial network, and respectively generating registered multi-mode MRI. The generatorin the generative adversarial network is modularized into the encoders and the decoders, through combined training of the encoders, the decoders and the discriminators, random input conforming to design specifications can be received, then a set of registered multi-mode MRI images with focus segmentation tags are generated, and the method can be widely applied to the field of medical images.

The invention discloses a solid insulation life prediction method based on transformer related operation data. The method comprises the following steps: 1) obtaining and storing transformer solid insulation life prediction related data; 2) reading related data for numbering; 3) preprocessing the data; 4) obtaining the closest distribution model by using a Monte Carlo method; 5) training the data by using the BP neural network to obtain a prediction model; 6) enabling the BP neural network to perform network initialization; 7) randomly selecting an adjusting parameter by adopting a Monte Carlomethod, and updating the loss function by adjusting a small amount of the parameter; and 8) updating the loss function of the solid insulation life prediction model through a Monte Carlo method to finally obtain life prediction of the line transformer solid insulation, thereby preventing the BP neural network from falling into the dilemma of a local minimum value, and meanwhile, having the advantages of simple structure and flexible training, and effectively improving the accuracy of transformer solid insulation evaluation.

The invention discloses a training management method, a training management device, electronic equipment and a computer readable storage medium. The training management method comprises the followingsteps of acquiring the current user grade of a user; inquiring the knowledge points to be learned of a user according to the correspondence table of a preset user grade and the knowledge points and the current user grade of the user; and generating an online learning path based on the knowledge points to be learnt, and training the user according to the online learning path. In the scheme of the invention, the online learning path suitable for the user can be made according to the current learning progress of the user so that the user can get targeted training and a training process becomes more flexible.

The invention discloses a gamification teaching platform and method and relates to the technical field of electronic teaching. Through an electronic game mode, working scenes, working actions, working relations and assessment evaluation required by engineering and integration courses are simulated. Defects of limited field equipment, incapability of ensuring the participation rate of students, lack of interestingness of practical training tasks, loss of objectivity in assessment and evaluation links and the like of engineering and integration courses at the present stage can be made up. The platform comprises a server, a teacher end and a student end, wherein the teacher end and the student end are in communication connection with the server to realize teaching information exchange; the server comprises a game task management module, a display module, a help seeking and rescuing module, an evaluation module, an award module, an attendance checking module and a user information management module; a teacher can randomly select scripts in the game task management module to generate different game tasks; the teacher end comprises a game task monitoring module; the student end comprises a game task implementation module.

The invention discloses a training learning method and system for a neural network of a paper pattern specification parameter inference model. The method comprises the steps that S1, all connection weights and thresholds are subjected to initialization setting after multiple groups of learning samples are imported into an input layer; S2, one group of learning samples are selected randomly, and input and output of all units in an interlayer are calculated through an interlayer algorithm; S3, output and response of all units of an output layer are calculated through an output layer algorithm; S4, an output layer generalized error of all the units of the output layer is calculated, and an interlayer generalized error of all the units of the interlayer is calculated through an interlayer generalized error algorithm; S5, the connection weight of the output layer is corrected, the output threshold of the output layer is corrected through a correction algorithm, the connection weight of the interlayer is corrected, and the output threshold of the interlayer is corrected through the correction algorithm; and S6, the next group of learning samples are selected randomly, and the step S2 is returned to. According to the training learning method and system, a network is converged step by step through selection of a proper training rate and other methods, and therefore the training learning effect is improved.

The embodiment of the invention provides a data prediction model training method and device and a storage medium. In the data prediction model training method, an appropriate model can be selected from a plurality of selectable prediction models according to a measurement error index and characteristics of historical data, and a corresponding optimal model is selected for predictable dependent variables and related independent variables needing to be predicted. A comprehensive prediction model may be trained based on the prediction model of the dependent variable and the relationship between the prediction model of the independent variable and the actual value of the dependent variable. The comprehensive prediction model can provide a universal prediction method, can flexibly implement prediction tasks related to various time sequences, and meets various prediction requirements of users.

The invention discloses a weight-bearing fitness facility and a weight-bearing operation method. The facility comprises a guide vertical frame, a guide transverse frame, a T-shaped frame, a transversesliding frame and a barbell rod. The guide vertical frame is fixedly arranged on the base in a vertical mode. Lifting oil cylinders are fixed to the two sides of the base correspondingly. Electromagnets are fixed at the middle positions of the upper part and the lower part of the guide cross frame; square sleeves are arranged on the two sides of the electromagnet. The T-shaped frame is inserted into the square sleeve and matched with the electromagnet, a transverse groove is formed in the guide transverse frame, the transverse sliding frame is in sliding fit with the transverse groove, the electromagnet is electrically connected with the handheld sleeve through a wire, the handheld sleeve is arranged on the barbell rod in a sleeving mode, and a switch is arranged on the handheld sleeve. The weight-bearing fitness facility is easy to operate, flexible to adjust, capable of meeting training of leg strength and waist strength actions, capable of being used independently without accompanying of other people when athletes act, high in operation safety and worthy of application and popularization, and the situation that the athletes get rid of strength and are injured is avoided.

The invention discloses a touch-screen multifunctional non-human primate animal cognitive-function testing method and belongs to the technical field of animal experiment devices. A touch screen capable of moving back and forth and a water supply nozzle capable of moving vertically are arranged in a testing cage. A monkey raised in a living cage is transferred to the testing cage through a transfer cage, the condition of the monkey in the transfer process is observed through an observation window of the transfer cage, a transfer cage door is in butt joint with a testing cage door after in-place transfer is completed, the transfer cage door and the testing cage door are pulled open, the testing cage door is closed through insertion after the monkey enters the testing cage, the monkey is enabled to complete a task at a natural sitting posture and obtain a drinking water award, an injector containing drinking water or fruit juice is accurately pushed through computer control, and the drinking water or fruit juice award is quantitatively given to the monkey. After the monkey correctly completes a screen touching task, the monkey drinks clear water or fruit juice from the water supply nozzle, and the quantitative relation formula of the height Y from the water supply nozzle to the bottom face of the testing cage and the distance X from the water supply nozzle to the touch face of the touch screen is determined as Y=0.4661X + (26-30), and measurement unit is millimeter.

The invention discloses a plantar driving type gait training device, which comprises a rack and more than one pair of motion training modules, and each motion training module comprises a track generating unit and a phase adjusting unit; the track generation unit comprises a differential gear train mechanism and a crank connecting rod mechanism, the differential gear train mechanism comprises an inner gear ring, a planetary gear, a planetary carrier and a driving shaft, and the planetary gear is located in the inner gear ring and engaged with the inner gear ring; the crank connecting rod mechanism comprises a crank and a connecting rod, and the crank is fixedly connected with the planetary gear; and the phase adjusting unit is used for driving the inner gear ring to rotate. The human body gait trajectory motion can be presented, the single training mode of the circular motion trajectory is changed, the form and size of the motion trajectory can be adjusted through the phase adjusting unit, the form is divided into an uphill mode, a downhill mode, a horizontal mode and a direct-acting mode, and the requirements of users with different heights and in different training stages are met.

The invention discloses a virtuality and reality combining data link networking simulation system and realization method. The virtuality and reality combining data link networking simulation system comprises a transmit-receive apparatus used for controlling the start of a data link analog module and setting networking parameters, a command and guide and target simulation module used for generating simulation commands, setting simulating target information and sending the simulating target information to a display and control module, the display and control module which comprises a picture organization module, a foreground display apparatus and a background logical processing module, and the data link simulation module used for connecting the display and control system and the transmit-receive apparatus of local equipment. According to the invention, the training cost is reduced, the simulation system and realization method can be realized flexibly, and the stability of the system is improved by adopting network communication.

According to the automatic ball passing control method and device for computer-aided training, the ball passing device can continuously lock a user and randomly transmit the balls to any position point in a circular area with the user as the circle center, the radius, the ball passing frequency and the ball passing mode of the circular area can be set, and different requirements of the user can bemet. According to the device, through the synergistic effect of the input module, the recognition tracking module, the projection module and the analysis control module, dynamic image recognition andtracking are achieved, a moving target is rapidly detected, the target is tracked, the ball passing direction and mode are controlled, and the device is not affected by indoor and outdoor environments and is suitable for professional training and training in other environments.

The invention relates to a shield tunneling machine operator training and evaluating system which comprises an operation training and evaluating unit, a three-dimensional simulation unit, an environment simulation and main control unit, a guide measurement unit and a comprehensive management system. The operation training and evaluating unit is used for being operated by an operator and generatinga corresponding operation signal. And the three-dimensional simulation unit is used for simulating the movement of a shield tunneling machine, displaying a three-dimensional model for simulating themovement state of the shield tunneling machine and driving the three-dimensional model to change according to the operation signal. And the environment simulation and main control unit is used for simulating and displaying the working environment of the shield tunneling machine, controlling the operation training and evaluating unit and receiving an operation signal. And the guide measurement unitis used for feeding back attitude data in the movement process of the shield tunneling machine in real time. According to the system, the shield tunneling machine operator can be trained and evaluated more flexibly and efficiently, the hardware cost can be greatly reduced; and the method has high economic benefits and popularization value.

The invention discloses a clinical adjustable limb rehabilitation apparatus for the neurology department, and relates to the technical field of auxiliary tools for the neurology department. According to the apparatus, the base is of a cavity structure, and a plurality of balancing weights are arranged in the base; two supporting plates are symmetrically arranged on the base; a first training device is arranged on the two supporting plates and comprises an arc-shaped sliding rail arranged between the two supporting plates, the two ends of the arc-shaped sliding rail are fixed to the side walls of the two supporting plates respectively, connecting holes are formed in the connecting positions of the two supporting plates and the arc-shaped sliding rail, and a limiting assembly is arranged on the arc-shaped sliding rail; two groups of hand training mechanisms are symmetrically arranged between the two supporting plates and the arc-shaped sliding rail; a seat is arranged on the base and located between the two supporting plates, and the bottom of the seat is connected with the base through a telescopic rod. The apparatus has the advantages that multiple different training modes are set to train muscle groups of the four limbs of a patient, the resistance assembly is further arranged to adjust resistance, the muscles of the patient can be well exercised, and recovery of the patient is facilitated.

The present invention relates to the field of deep learning image generation, and in particular to a multi-domain image conversion method, system and medium based on conditional generative confrontation networks. Original image y; use a pre-trained condition extractor to produce x modal condition C x and the y-mode condition C y ;The original image x, original image y, x modal condition C x , y modality inputs the pre-trained conditional generative adversarial network to obtain the corresponding image conversion result. The present invention uses a feature extractor to extract the features of the original image, obtains a conditional matrix through upsampling and splicing with the zero matrix on the channel, and retains the semantic information of each modal input in the case of high independence; The invention is flexible in training, has no limit to the number of domains to be converted, and requires few parameters.

The invention provides a high-risk chemical laboratory fireproof safety training method based on VR and artificial intelligence. The method comprises the steps: step 1, establishing a virtual scene for high-risk chemical laboratory fireproof safety training; and step 2, in the virtual scene, picking up a virtual object and touching the virtual object by using voice recognition, and recognizing theidentity of a trainee by using voiceprint recognition so as to complete high-risk chemical laboratory fireproof safety training. According to the invention, the virtual scene is constructed; voice recognition and voiceprint recognition are combined with the virtual scene to carry out high-risk chemical laboratory fireproof safety training, the trainee as a participant of VR fire-fighting drillingcan personally experience operations such as fire extinguishing, escape and self-rescue in a fire disaster, and the fire-fighting knowledge and skills of an experiencer are enhanced through realistically restoring a fire scene.

The invention is directed to a computer-implemented method for determining at least one completed item of at least one product solution, comprising the steps of: a. Providing at least one input data set with at least one partial item of the at least one product solution; wherein b. the at least one partial item comprises at least one initial feature; c. Complementing the at least one partial item of the at least one product solution with at least one additional alternative feature using a trained machine learning model on the basis of at least one partial item of the at least one product solution to determine a plurality of alternative complete items of the at least one product solution; and d. Determining at least one evaluated complete item of the plurality of alternative items of the at least one product solution as output data set using a market impact evaluation. Further, the invention relates to a corresponding computer program product and system.

Provided is a computer-implemented method for determining at least one class, including the steps of: providing at least one input data set with a plurality of performance metrics; preprocessing the at least one input data set into at least one respective processed input data set with a plurality of processed performance metrics; and determining the at least one class using machine learning on the basis of the at least one processed input data set. Further, a corresponding computer program product and system is provided.

The invention discloses a wearable glasses with variable prismatic degrees. A patient directly wears the glasses into which different prisms can be inserted to carry out the free space visual training, the bottom outward training collection of two eyes and the bottom inward training divergence. According to the technical scheme, the glasses comprises a glasses frame, two glasses frames and prismatic lenses of various specifications, first lens grooves used for embedding common lenses are formed in the glasses frames, second lens grooves used for horizontally placing the prismatic lenses are formed in the outer sides of the glasses frames, and positioning frames are arranged at the peripheries of the prismatic lenses and can be embedded into the second lens grooves. The glasses of the present invention is not limited by space places and instruments, the patient can wear the glasses to freely walk to carry out visual training, the objective images are displaced through the refraction effect of the prisms, and human eyes need to actively gather or diverge to maintain the binocular single vision. The prisms with different prism degrees are installed in the glasses frame, the prism-wearing training requires the human eyes to always adapt to and use divergence or gathering and maintain a certain tension level, so that a continuous training effect is achieved.

The invention discloses a data link network simulation system combined with virtual reality and an implementation method thereof. The data link network simulation system combined with virtual reality includes: a transceiver device, which is used to start control the data link simulation module and control the network Parameters are set; the command guidance and target simulation module is used to generate simulation commands, and is used to set simulated target information and send it to the display and control module; the display and control module includes a screen organization module, a front display device, and a background logic processing module ; The data link simulation module is used to connect the display control system and the transceiver device of the local equipment. The simulation system and the realization method of the invention reduce the training cost, are flexible in realization, and the stability of the network communication system is also enhanced.

The invention belongs to the technical field of information retrieval and its database structure, and discloses a semantic calculation-based classifier construction method and a classifier, which use a neural network model to train word vectors on Wikipedia data to obtain distributed representations of words; through classification The vector representation of the classification is obtained by the label, and the vector representation of the text is obtained by using the weighted average method; the most likely classification of the text is obtained by calculating the semantic relationship between the classification vector and the text vector. The unsupervised learning stage of the present invention does not need to label data, only needs the user to define a small number of feature words to complete the creation of the classifier, the online speed is fast, and there is no need to wait for the long accumulation of label data; the unsupervised learning stage can make full use of the existing limited labels Data, guide and improve unsupervised classification ability by extracting effective feature words.

The invention discloses a touch-screen multifunctional non-human primate animal cognitive-function testing method and belongs to the technical field of animal experiment devices. A touch screen capable of moving back and forth and a water supply nozzle capable of moving vertically are arranged in a testing cage. A monkey raised in a living cage is transferred to the testing cage through a transfer cage, the condition of the monkey in the transfer process is observed through an observation window of the transfer cage, a transfer cage door is in butt joint with a testing cage door after in-place transfer is completed, the transfer cage door and the testing cage door are pulled open, the testing cage door is closed through insertion after the monkey enters the testing cage, the monkey is enabled to complete a task at a natural sitting posture and obtain a drinking water award, an injector containing drinking water or fruit juice is accurately pushed through computer control, and the drinking water or fruit juice award is quantitatively given to the monkey. After the monkey correctly completes a screen touching task, the monkey drinks clear water or fruit juice from the water supply nozzle, and the quantitative relation formula of the height Y from the water supply nozzle to the bottom face of the testing cage and the distance X from the water supply nozzle to the touch face of the touch screen is determined as Y=0.4661X + (26-30), and measurement unit is millimeter.

The invention provides a training and fitness device with a residual force utilization function based on osteoporosis prevention, and relates to the technical field of fitness equipment. The training and fitness device comprises a supporting device, wherein two arm strength training parts are mounted on the supporting device; the bottom of the supporting device is slidably connected with a linkage part; the bottom of the supporting device is fixedly connected with a driving part; the linkage part is fixedly connected with four floating massage parts; the linkage part is fixedly connected with a hip supporting device, and the hip supporting device is slidably connected to the supporting device; and the hip supporting device is rotatably connected with two joint prevention parts, and the two joint prevention parts are slidably connected with blood circulation parts respectively, so that the use is more flexible, the osteoporosis prevention effect is better, training is safer and more relaxed, the training efficiency is improved, and the problem that an existing training and fitness device is large in training intensity, not capable of conducting protective massage, easy to cause secondary injury, low in training comfort and poor in training effect is solved.