unreal.MLLevelSetModelInferenceInfo

class unreal.MLLevelSetModelInferenceInfo(nne_model_path: str = '', model_architecture_activation_node_sizes: None = [], ml_model_weights: str = '')

Bases: TableRowBase

MLLevel Set Model Inference Info

C++ Source:

  • Module: Engine

  • File: MLLevelSetModelAndBonesBinningInfo.h

Editor Properties: (see get_editor_property/set_editor_property)

  • ml_model_weights (str): [Read-Write] Model weights {W1,..,Wn} is tokanized into FString as “W1_0,W1_1,…,W1_k1|W2_0,W2_1,…,W2_k2|…|Wn_0,Wn_1,…,Wn_kn”

  • model_architecture_activation_node_sizes (Array[int32]): [Read-Write] In most cases input layer is 3 (location input, X,Y,Z coordinates of the query), and output layer is 1 (Scaled Signed Distance or Incorrect Zone Indicator)

  • nne_model_path (str): [Read-Write] The path to the NNE model. At the moment MLLevelSet Asset accepts only mlir.tosa models.

property ml_model_weights: str

[Read-Write] Model weights {W1,..,Wn} is tokanized into FString as “W1_0,W1_1,…,W1_k1|W2_0,W2_1,…,W2_k2|…|Wn_0,Wn_1,…,Wn_kn”

Type:

(str)

property model_architecture_activation_node_sizes: None

[Read-Write] In most cases input layer is 3 (location input, X,Y,Z coordinates of the query), and output layer is 1 (Scaled Signed Distance or Incorrect Zone Indicator)

Type:

(Array[int32])

property nne_model_path: str

[Read-Write] The path to the NNE model. At the moment MLLevelSet Asset accepts only mlir.tosa models.

Type:

(str)