Deep Learning Model Package Inference
|
From Mech-Vision 2.2.0, this Step only supports loading model packages exported from Mech-DLK 2.5.4 or above. |
|
If you need to determine the exported version of an existing model package, you can open the ".dlkpack" file with 7-Zip or other decompression software:
|
Function
This Step performs inference with single model packages or multiple model packages exported from Mech-DLK and outputs the inference result.
|
In Mech-Vision 2.2.0, if the “Deep Learning Model Package Inference” Step loads a model package exported from Mech-DLK 3.0.0 or earlier that contains modules with configured filtering rules, these filtering rules will be cleared. In this case, please reconfigure and export the model package in Mech-DLK 3.0.0 or above, and then use it in the Deep Learning Model Package Inference Step. |
Usage Scenario
Currently, this Step can perform model package inference in the following scenarios: classification, object detection, defect segmentation, instance segmentation, fast positioning, text detection, text recognition, unsupervised segmentation, and multi-model packages. For a description of the compatibility of the Step, please refer to Compatibilities of Deep Learning Steps.
Input and Output
The input and output ports of this Step vary according to the selected model package type. Please select the corresponding model package type below to view the changes.
System Requirements
The following system requirements need to be met when using this Step.
-
CPU: needs to support the AVX2 instruction set and meets any of the following conditions:
-
IPC or PC without any discrete graphics card: Intel i5-12400 or higher.
-
IPC or PC with a discrete graphics card: Intel i7-6700 or higher, with the graphics card not lower than GeForce GTX 1660.
This Step has been thoroughly tested on Intel CPUs but has not been tested on AMD CPUs yet. Therefore, Intel CPUs are recommended.
-
-
GPU: GeForce GTX 1660 or above (if with a discrete graphics card).
|
Ensure that the minimum required version for the GPU driver is 526.98, and the minimum required CPU version is 6th-generation Intel Core. When the hardware cannot meet the requirement, the deep learning model package cannot be imported successfully. |
Select Model Package Type
You can click the links below to view the function description, input and output, and parameter description of each model package type.