Stay organized with collections
    
    
      
      Save and categorize content based on your preferences.
    
  
  
    
  
  
  
  
    
  
  
    
    
tensorflow::tensorrt::TfTrtConversionParams
Summary
Public attributes
allow_build_at_runtime
bool tensorflow::tensorrt::TfTrtConversionParams::allow_build_at_runtime = true
 
convert_to_static_engine
bool tensorflow::tensorrt::TfTrtConversionParams::convert_to_static_engine = true
 
max_cached_engines
int tensorflow::tensorrt::TfTrtConversionParams::max_cached_engines = 1
 
max_workspace_size_bytes
size_t tensorflow::tensorrt::TfTrtConversionParams::max_workspace_size_bytes = 1 << 30
 
minimum_segment_size
int tensorflow::tensorrt::TfTrtConversionParams::minimum_segment_size = 3
 
precision_mode
TrtPrecisionMode tensorflow::tensorrt::TfTrtConversionParams::precision_mode = TrtPrecisionMode::FP32
 
profile_strategy
ProfileStrategy tensorflow::tensorrt::TfTrtConversionParams::profile_strategy = ProfileStrategy::kRange
 
use_calibration
bool tensorflow::tensorrt::TfTrtConversionParams::use_calibration = true
 
use_dynamic_shape
bool tensorflow::tensorrt::TfTrtConversionParams::use_dynamic_shape = true
 
  
  
 
  
    
    
      
       
    
    
  
  
  Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates. Some content is licensed under the numpy license.
  Last updated 2022-02-17 UTC.
  
  
  
    
      [null,null,["Last updated 2022-02-17 UTC."],[],[]]