QuantizedTeacherModel<T>
Models & Types ยท AiDotNet.KnowledgeDistillation.Teachers
Quantized teacher model with reduced precision for efficient deployment.
Inheritance: TeacherModelBase<Vector<T>, Vector<T>, T> โ QuantizedTeacherModel<T>
For Beginners
Quantization reduces the numerical precision of model weights and activations to use fewer bits (e.g., 8-bit instead of 32-bit floating point). This enables:
Constructors
| Constructor | Summary |
|---|---|
QuantizedTeacherModel(ITeacherModel<Vector<T>, Vector<T>> baseTeacher, int quantizationBits) |
Initializes a new instance of QuantizedTeacherModel wrapping a teacher interface. |
Properties
| Property | Summary |
|---|---|
int OutputDimension { get; } |
Gets the output dimension of the teacher model. |
Methods
| Method | Summary |
|---|---|
Vector<T> GetLogits(Vector<T> input) |
Gets quantized logits from the teacher model. |