4 results about "Operating model" patented technology

A closed dialog-based input system (200) for generating a binding operating model with a fixed execution reference, comprising: a stepwise dialog acquisition unit (210) with a receiving interface for successive dialog-based inputs via any physical or logical input channel, wherein separate dialog stages are storable and referenceable; a dialog state fixing unit (220) for fixing an operational state after each dialog stage; and a dialog conversion unit (230) with an extraction unit for a predefined set of operational parameters from each fixed dialog stage, as well as a generation unit for referential operational parameters in the absence of explicit input of certain parameters.- a unit for generating a fixed dialogic and execution-related reference as well as a unique identifier of an independent execution cycle (240), - a unit for structural dialogic validation (250) with a validation structure for ensuring the reference consistency of all dialogic stages and operational parameters within the execution cycle, - a dialogic execution lock (260), and - a stepwise dialog and execution protocol (270), characterized in that the input system (200) is configured such that - after each dialogic stage, a fixed operational state data record is generated and stored in a reference-bound manner, and - the system architecture technically blocks a transition to a subsequent dialogic stage as well as the transfer of operational parameters or the generation of operational outputs before all operational parameters are extracted or generated, fixed,bound to the fixed execution reference, structurally validated and fixed in a final dialog and execution protocol (270) with a single unchanging reference.

The invention provides a model compiling method and device. The method comprises the following steps: determining a hardware feature vector of target hardware and analysis data of a to-be-operated model on the target hardware; based on the hardware feature vector and the analysis data of the to-be-operated model, generating a first optimization strategy and a second optimization strategy by using a decision rule and a machine learning model; fusing the first optimization strategy and the second optimization strategy to obtain a target optimization strategy; and under the condition that the confidence evaluation result of the target optimization strategy meets the requirement, the target optimization strategy is applied to the compiler, so that the compiler compiles the to-be-operated model according to the target optimization strategy.

The application belongs to the field of industrial internet, and particularly relates to a kind of intelligent interaction three-dimensional modeling method, device, equipment and storage medium. First, in response to the interaction operation triggered by the client, the modeling text is determined, then the workpiece to be assembled and its three-dimensional modeling parameters are determined according to the modeling text, the corresponding three-dimensional modeling template is found out and spliced with the parameters to generate a three-dimensional modeling scheme, then the scheme is input into the preset modeling model to obtain a three-dimensional target model, and finally the model is output. The method automatically analyzes the modeling text and completes three-dimensional modeling with the help of the preset template and model, solves the problem that non-professional users cannot operate modeling software for modeling, realizes efficient and convenient three-dimensional modeling, and reduces the modeling threshold.

The invention provides a GPU resource scheduling method and device. According to the embodiment of the invention, the feedback of the user is used as a standard for determining whether the model can obtain resources, and more users can enjoy high-quality reasoning service through flexible and dynamic resource adjustment. Besides, GPU resources (such as peak-valley allocation, elastic capacity reduction and the like) are more scientifically and reasonably scheduled to a to-be-operated model task according to the GPU resources required by the new model during training and reasoning in combination with sorting and peak-valley time period judgment of the deployed model, so that the GPU resources are ensured to be obtained by more tasks, the resource utilization rate of the GPU is greatly improved, and the resource utilization rate of the new model is increased. And the GPU purchase cost of a client (a model service provider) is indirectly reduced.