Module inference

Module inference 

Source
Expand description

Inference optimization utilities

This module provides optimizations for model inference including:

  • Operator fusion
  • Constant folding
  • Memory optimization
  • Batch inference

Structs§

BatchInference
Batch inference helper
FusedOp
Fused operation
InferenceConfig
Inference mode configuration
InferenceOptimizer
Inference optimizer
InferenceSession
Inference session for optimized model execution

Functions§

warmup_model
Warmup helper for inference