inference-engine/samples/hello_query_device/README.md

   1 # Hello Query Device C++ Sample
   2
   3 This topic demonstrates how to run the Hello Query Device sample application, which queries Inference Engine devices and prints their metrics and default configuration values. The sample shows how to use [Query Device API feature](./docs/IE_DG/InferenceEngine_QueryAPI.md).
   4 > **NOTE:** This topic describes usage of C++ implementation of the Query Device Sample.
   5 > For the Python* implementation, refer to [Hello Query Device Python* Sample](./inference-engine/ie_bridges/python/sample/hello_query_device/README.md)
   6 ## Running
   7
   8 To see quired information, run the following:
   9 ```sh
  10 ./hello_query_device
  11 ```
  12
  13 ## Sample Output
  14
  15 The application prints all available devices with their supported metrics and default values for configuration parameters:
  16
  17 ```
  18 Available devices:
  19         Device: CPU
  20         Metrics:
  21                 AVAILABLE_DEVICES : [ 0 ]
  22                 SUPPORTED_METRICS : [ AVAILABLE_DEVICES SUPPORTED_METRICS FULL_DEVICE_NAME OPTIMIZATION_CAPABILITIES SUPPORTED_CONFIG_KEYS RANGE_FOR_ASYNC_INFER_REQUESTS RANGE_FOR_STREAMS ]
  23                 FULL_DEVICE_NAME : Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz
  24                 OPTIMIZATION_CAPABILITIES : [ WINOGRAD FP32 INT8 BIN ]
  25                 SUPPORTED_CONFIG_KEYS : [ CPU_BIND_THREAD CPU_THREADS_NUM CPU_THROUGHPUT_STREAMS DUMP_EXEC_GRAPH_AS_DOT DYN_BATCH_ENABLED DYN_BATCH_LIMIT EXCLUSIVE_ASYNC_REQUESTS PERF_COUNT ]
  26                 ...
  27         Default values for device configuration keys:
  28                 CPU_BIND_THREAD : YES
  29                 CPU_THREADS_NUM : 0
  30                 CPU_THROUGHPUT_STREAMS : 1
  31                 DUMP_EXEC_GRAPH_AS_DOT : ""
  32                 DYN_BATCH_ENABLED : NO
  33                 DYN_BATCH_LIMIT : 0
  34                 EXCLUSIVE_ASYNC_REQUESTS : NO
  35                 PERF_COUNT : NO
  36
  37         Device: FPGA
  38         Metrics:
  39                 AVAILABLE_DEVICES : [ 0 ]
  40                 SUPPORTED_METRICS : [ AVAILABLE_DEVICES SUPPORTED_METRICS SUPPORTED_CONFIG_KEYS FULL_DEVICE_NAME OPTIMIZATION_CAPABILITIES RANGE_FOR_ASYNC_INFER_REQUESTS ]
  41                 SUPPORTED_CONFIG_KEYS : [ DEVICE_ID PERF_COUNT EXCLUSIVE_ASYNC_REQUESTS DLIA_IO_TRANSFORMATIONS_NATIVE DLIA_ARCH_ROOT_DIR DLIA_PERF_ESTIMATION ]
  42                 FULL_DEVICE_NAME : a10gx_2ddr : Intel Vision Accelerator Design with Intel Arria 10 FPGA (acla10_1150_sg10)
  43                 OPTIMIZATION_CAPABILITIES : [ FP16 ]
  44                 RANGE_FOR_ASYNC_INFER_REQUESTS : { 2, 5, 1 }
  45         Default values for device configuration keys:
  46                 DEVICE_ID : [ 0 ]
  47                 PERF_COUNT : true
  48                 EXCLUSIVE_ASYNC_REQUESTS : false
  49                 DLIA_IO_TRANSFORMATIONS_NATIVE : false
  50                 DLIA_PERF_ESTIMATION : true
  51 ```
  52
  53 ## See Also
  54 * [Using Inference Engine Samples](./docs/IE_DG/Samples_Overview.md)
  55 * [Model Downloader](https://github.com/opencv/open_model_zoo/tree/2018/model_downloader)
  56 * [Model Optimizer](./docs/MO_DG/Deep_Learning_Model_Optimizer_DevGuide.md)