OnnxLayerOutputs
Models & Types · AiDotNet.Onnx
Named tensors flowing OUT of a layer’s ONNX node(s). The names are the tensors a
layer’s ONNX nodes write to; downstream layers will reference them as inputs.
The graph builder uses these names to wire the next layer’s OnnxLayerInputs.
Implements: IEquatable<OnnxLayerOutputs>
Constructors
| Constructor | Summary |
|---|---|
OnnxLayerOutputs(IReadOnlyList<string> TensorNames) |
Convenience constructor for the common single-output case. |
OnnxLayerOutputs(string tensorName) |
Convenience constructor for the common single-output case. |
Properties
| Property | Summary |
|---|---|
IReadOnlyList<string> TensorNames { get; set; } |
|
bool HasMultiple { get; } |
True if this layer produces more than one output tensor (e.g., a split or a BN training-mode emit). |
string Primary { get; } |
The primary output tensor name (the first entry). |
Methods
| Method | Summary |
|---|---|
OnnxLayerOutputs <Clone>$() |
|
bool Equals(OnnxLayerOutputs other) |
|
bool Equals(object obj) |
|
bool op_Equality(OnnxLayerOutputs left, OnnxLayerOutputs right) |
|
bool op_Inequality(OnnxLayerOutputs left, OnnxLayerOutputs right) |
|
int GetHashCode() |
|
string ToString() |
|
void Deconstruct(IReadOnlyList<string> TensorNames) |