Change that default value is one of MV_INFERENCE_DATA_TYPE 59/234059/2 accepted/tizen/unified/20200603.150123 submit/tizen/20200602.011936 submit/tizen/20200602.084841
authorTae-Young Chung <ty83.chung@samsung.com>
Thu, 21 May 2020 09:48:57 +0000 (18:48 +0900)
committerTae-Young Chung <ty83.chung@samsung.com>
Fri, 22 May 2020 02:31:41 +0000 (11:31 +0900)
MV_INFERENCE_DATA_TYPE supports MV_INFERENCE_DATA_FLOAT32 and
MV_INFERENCE_DATA_UINT8. But in media-vision-config.json,
its default value is '-1'.
It should be one of MV_INFERENCE_DATA_TYPE.
In addition, add the description of MV_INFERENCE_DATA_TYPE

Change-Id: I5debd73de3073715657aed25405d5098bf41a0ed
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
include/mv_inference.h
media-vision-config.json

index 97966d321ebe63084431393b56bf4b02467310b5..878c19bef4c052f9ce8823287598b004e47812b7 100644 (file)
@@ -170,6 +170,11 @@ extern "C" {
 /**
  * @brief Defines #MV_INFERENCE_INPUT_DATA_TYPE to set data type of input tensor.
  * @details Data type of input tensor can be changed according to a given weight file.
+ *          Switches between Float32 or UInt8:\n
+ *          #MV_INFERENCE_DATA_FLOAT32,\n
+ *          #MV_INFERENCE_DATA_UINT8,\n
+ *
+ *          The default type is MV_INFERENCE_DATA_FLOAT32.
  *
  * @since_tizen 6.0
  * @see mv_engine_config_set_int_attribute()
index 2116fb6b2a6e2594d57fc049bef9e3f2326da222..59dbeaf00ec8bfba4de1571d161d29cf07a2219f 100644 (file)
         {
             "name" : "MV_INFERENCE_INPUT_DATA_TYPE",
             "type" : "integer",
-            "value" : -1
+            "value" : 0
         },
         {
             "name"  : "MV_INFERENCE_MODEL_CONFIGURATION_FILE_PATH",