require_full_delegation=false bool optional require delegate to run the entire graph allow_fp16=false bool optional allow fp16 The file format is binary and it should be array format or null separated strings format. If the input_name appears both in input_layer_value_range and input_layer_value_files, input_layer_value_range of the input_name will be ignored. In case the input layer name contains ':' e.g. Each item is separated by ',', and the item value consists of input layer name and value file path separated by ':', e.g. input_layer_value_files= string optionalĚ map-like string representing value file. Each item is separated by ':', and the item value consists of input layer name and integer-only range values (both low and high are inclusive) separated by ',', e.g. input_layer_value_range= string optionalĚ map-like string representing value range for *integer* input layers. input_layer_shape= string optional input layer shape input_layer= string optional input layer names This is only used when -report_peak_memory_footprint is set to true. memory_footprint_check_interval_ms=50 int32 optional The interval in millisecond between two consecutive memory footprint checks. Therefore, the performance benchmark result could be affected. Internally, a separate thread will be spawned for this periodic check. report_peak_memory_footprint=false bool optional Report the peak memory footprint by periodically checking the memory footprint. but without actually invoking any op kernels. dry_run=false bool optional Whether to run the tool just with simply loading the model, allocating tensors etc. By default, only log those parameters that are set by parsing their values from the commandline flags. verbose=false bool optional Whether to log parameters whose values are not set. warmup_min_secs=0.5 float optional minimum number of seconds to rerun for, potentially making the actual number of warm-up runs to be greater than warmup_runs warmup_runs=1 int32 optional minimum number of runs performed on initialization, to allow performance characteristics to settle, see also warmup_min_secs output_prefix= string optional benchmark output prefix benchmark_name= string optional benchmark name Currently implies the use of the Ruy library. use_caching=false bool optionalĞnable caching of prepacked weights matrices in matrix multiplication routines. num_threads=-1 int32 optional number of threads run_frequency=-1 float optionalĞxecute at a fixed frequency, instead of a fixed delay.Note if the targeted rate per second cannot be reached, the benchmark would start the next run immediately, trying its best to catch up. run_delay=-1 float optional delay between runs in seconds Note if -max-secs is exceeded in the middle of a run, the benchmark will continue to the end of the run but will not start the next run. max_secs=150 float optional maximum number of seconds to rerun for, potentially making the actual number of runs to be less than num_runs. min_secs=1 float optional minimum number of seconds to rerun for, potentially making the actual number of runs to be greater than num_runs num_runs=50 int32 optional expected number of runs, see also min_secs, max_secs It accepts the following input parameters: usr/local/bin/tensorflow-lite-x.x.x/tools/benchmark_model The benchmark_model C/C++ application is located in the userfs partition: The model used in this example can be installed from the following package:Īpt-get install tflite-models-mobilenetv1Ģ How to use the Benchmark application 2.1 Executing with the command line The detailed content licenses can be found here.Īfter having configured the AI OpenSTLinux package install X-LINUX- AI components for this application. The software package is provided AS IS, and by downloading it, you agree to be bound to the terms of the software license agreement (SLA). 1.1 Installing from the OpenSTLinux AI package repositoryġ Installation 1.1 Installing from the OpenSTLinux AI package repository Warning.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |