The NNStreamer Single API provides interfaces to invoke a neural network model with a single instance of input data.

Required Header

#include <nnstreamer/nnstreamer-single.h>

Overview

The NNStreamer Single API provides interfaces to invoke a neural network model with a single instance of input data. This function is a syntactic sugar of NNStreamer Pipeline API with simplified features; thus, users are supposed to use NNStreamer Pipeline API directly if they want more advanced features. The user is expected to preprocess the input data for the given neural network model.

This function allows the following operations with NNStreamer:

Open a machine learning model with various mechanisms.
Close the model.
Interfaces to enter a single instance of input data to the opened model.
Utility functions to get the information of opened model.

Note that this function set is supposed to be thread-safe.

Related Features

This function is related with the following features:

http://tizen.org/feature/machine_learning
http://tizen.org/feature/machine_learning.inference

It is recommended to probe features in your application for reliability.
You can check if a device supports the related features for this function by using System Information, thereby controlling the procedure of your application.
To ensure your application is only running on the device with specific features, please define the features in your manifest file using the manifest editor in the SDK.
For example, your application accesses to the camera device, then you have to add 'http://tizen.org/privilege/camera' into the manifest of your application.
More details on featuring your application can be found from Feature Element.

Functions
int	ml_single_open (ml_single_h single, const char model, const ml_tensors_info_h input_info, const ml_tensors_info_h output_info, ml_nnfw_type_e nnfw, ml_nnfw_hw_e hw)
	Opens an ML model and returns the instance as a handle.
int	ml_single_open_full (ml_single_h single, const char model, const ml_tensors_info_h input_info, const ml_tensors_info_h output_info, ml_nnfw_type_e nnfw, ml_nnfw_hw_e hw, const char *custom_option)
	Opens an ML model and returns the instance as a handle with custom options.
int	ml_single_close (ml_single_h single)
	Closes the opened model handle.
int	ml_single_invoke (ml_single_h single, const ml_tensors_data_h input, ml_tensors_data_h *output)
	Invokes the model with the given input data.
int	ml_single_invoke_fast (ml_single_h single, const ml_tensors_data_h input, ml_tensors_data_h output)
	Invokes the model with the given input data and fills the output data handle. This is generally faster than ml_single_invoke().
int	ml_single_invoke_dynamic (ml_single_h single, const ml_tensors_data_h input, const ml_tensors_info_h in_info, ml_tensors_data_h output, ml_tensors_info_h out_info)
	Invokes the model with the given input data with the given tensors information.
int	ml_single_get_input_info (ml_single_h single, ml_tensors_info_h *info)
	Gets the information (tensor dimension, type, name and so on) of required input data for the given model.
int	ml_single_get_output_info (ml_single_h single, ml_tensors_info_h *info)
	Gets the information (tensor dimension, type, name and so on) of output data for the given model.
int	ml_single_set_input_info (ml_single_h single, const ml_tensors_info_h info)
	Sets the information (tensor dimension, type, name and so on) of required input data for the given model.
int	ml_single_set_timeout (ml_single_h single, unsigned int timeout)
	Sets the maximum amount of time to wait for an output, in milliseconds.
int	ml_single_set_property (ml_single_h single, const char name, const char value)
	Sets the property value for the given model.
int	ml_single_get_property (ml_single_h single, const char name, char *value)
	Gets the property value for the given model.
int	ml_single_open_with_option (ml_single_h *single, const ml_option_h option)
	Makes a single instance with given ml-option.
Typedefs
typedef void *	ml_single_h
	A handle of a single-shot instance.

Typedef Documentation

typedef void* ml_single_h

A handle of a single-shot instance.

Since :: 5.5

Function Documentation

int ml_single_close ( ml_single_h single )

Closes the opened model handle.

Note that this should be called before destroying the inference data by ml_tensors_data_destroy(). If not, the inference engine might try to access the data that is already freed. And it causes the segmentation fault.

Since :: 5.5

Parameters:

[in] single The model handle to be closed.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid (Pipeline is not negotiated yet.)

int ml_single_get_input_info	(	ml_single_h	single,
		ml_tensors_info_h *	info
	)

Gets the information (tensor dimension, type, name and so on) of required input data for the given model.

Note that a model may not have such information if its input type is flexible. The names of tensors are sometimes unavailable (optional), while its dimensions and types are always available.

Since :: 5.5

Remarks:: The info should be released using ml_tensors_info_destroy().

Parameters:

[in]	single	The model handle.
[out]	info	The handle of input tensors information.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid. ml_tensors_info_h info = NULL; ml_single_get_input_info (single, &info); ... do something .... ml_tensors_info_destroy (info);

int ml_single_get_output_info	(	ml_single_h	single,
		ml_tensors_info_h *	info
	)

Gets the information (tensor dimension, type, name and so on) of output data for the given model.

Note that a model may not have such information if its output type is flexible and output type is not determined statically. The names of tensors are sometimes unavailable (optional), while its dimensions and types are always available.

Since :: 5.5

Remarks:: The info should be released using ml_tensors_info_destroy().

Parameters:

[in]	single	The model handle.
[out]	info	The handle of output tensors information.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid. ml_tensors_info_h info = NULL; ml_single_get_output_info (single, &info); ... do something .... ml_tensors_info_destroy (info);

int ml_single_get_property	(	ml_single_h	single,
		const char *	name,
		char **	value
	)

Gets the property value for the given model.

Since :: 6.0

Remarks:: The value should be released using g_free().

Parameters:

[in]	single	The model handle.
[in]	name	The property name.
[out]	value	The property value.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid. char *value; ml_single_get_property (single, "prop_name", &value); ... do something .... g_free (value);

int ml_single_invoke	(	ml_single_h	single,
		const ml_tensors_data_h	input,
		ml_tensors_data_h *	output
	)

Invokes the model with the given input data.

Even if the model has flexible input data dimensions, input data frames of an instance of a model should share the same dimension. Note that this will wait for the result until the invoke process is done. If an application wants to change the time to wait for an output, set the timeout using ml_single_set_timeout().

Since :: 5.5

Remarks:: The output should be released using ml_tensors_data_destroy().

Parameters:

[in]	single	The model handle to be inferred.
[in]	input	The input data to be inferred.
[out]	output	The allocated output buffer.

Returns:: 0 on success. Otherwise a negative error value.

Note:: If the data for the output buffer is allocated by the neural network framework (ML_NNFW_TYPE_CUSTOM_FILTER supports this), then this buffer will be freed when closing the single automatically by the neural network framework, and will not available for use later. It is recommended to copy the output buffer from output if it is required to use it after the single handle is closed.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid.
ML_ERROR_STREAMS_PIPE	Failed to push a buffer into source element.
ML_ERROR_TIMED_OUT	Failed to get the result from sink element.
ML_ERROR_OUT_OF_MEMORY	Failed to allocate required memory. ml_tensors_data_h output = NULL; ml_single_invoke (single,input, &output); ... do something .... ml_tensors_data_destroy (output);

int ml_single_invoke_dynamic	(	ml_single_h	single,
		const ml_tensors_data_h	input,
		const ml_tensors_info_h	in_info,
		ml_tensors_data_h *	output,
		ml_tensors_info_h *	out_info
	)

Invokes the model with the given input data with the given tensors information.

This function changes the input tensors information for the model, and returns the corresponding output data. A model/framework may not support changing the information. Note that this will wait for the result until the invoke process is done. If an application wants to change the time to wait for an output, set the timeout using ml_single_set_timeout().

Since :: 6.0

Remarks:: The output should be released using ml_tensors_data_destroy().; The out_info should be released using ml_tensors_info_destroy().

Parameters:

[in]	single	The model handle to be inferred.
[in]	input	The input data to be inferred.
[in]	in_info	The handle of input tensors information.
[out]	output	The allocated output buffer.
[out]	out_info	The handle of output tensors information.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid.
ML_ERROR_STREAMS_PIPE	Failed to push a buffer into source element.
ML_ERROR_TIMED_OUT	Failed to get the result from sink element.
ML_ERROR_OUT_OF_MEMORY	Failed to allocate required memory. ml_tensors_data_h output = NULL; ml_tensors_info_h out_info = NULL; ml_single_invoke_dynamic (single, input, in_info, &output, &out_info); ... do something .... ml_tensors_data_destroy (output); ml_tensors_info_destroy (out_info);

int ml_single_invoke_fast	(	ml_single_h	single,
		const ml_tensors_data_h	input,
		ml_tensors_data_h	output
	)

Invokes the model with the given input data and fills the output data handle. This is generally faster than ml_single_invoke().

The caller should preallocate memory buffers of the given output handle before calling the API. Note that ml_single_invoke() allocates memory buffers of the output handle in the API, which may incur memcpy.

Since :: 6.5

Parameters:

[in]	single	The model handle to be inferred.
[in]	input	The input data to be inferred.
[in,out]	output	The output data to be filled by the API. Output should be preallocated before calling the API.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid.
ML_ERROR_STREAMS_PIPE	Failed to push a buffer into source element.
ML_ERROR_TIMED_OUT	Failed to get the result from sink element.
ML_ERROR_OUT_OF_MEMORY	Failed to allocate required memory.

int ml_single_open	(	ml_single_h *	single,
		const char *	model,
		const ml_tensors_info_h	input_info,
		const ml_tensors_info_h	output_info,
		ml_nnfw_type_e	nnfw,
		ml_nnfw_hw_e	hw
	)

Opens an ML model and returns the instance as a handle.

Even if the model has flexible input data dimensions, input data frames of an instance of a model should share the same dimension.

Since :: 5.5

Remarks:: http://tizen.org/privilege/mediastorage is needed if model is relevant to media storage.; http://tizen.org/privilege/externalstorage is needed if model is relevant to external storage.; The single should be released using ml_single_close().

Parameters:

[out]	single	This is the model handle opened.
[in]	model	This is the path to the neural network model file.
[in]	input_info	This is required if the given model has flexible input dimension, where the input dimension MUST be given before executing the model. It is required by some custom filters of NNStreamer. You may set NULL if it's not required.
[in]	output_info	This is required if the given model has flexible output dimension.
[in]	nnfw	The neural network framework used to open the given model. Set ML_NNFW_TYPE_ANY to let it auto-detect.
[in]	hw	Tell the corresponding nnfw to use a specific hardware. Set ML_NNFW_HW_ANY if it does not matter.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_PERMISSION_DENIED	The application does not have the privilege to access to the media storage or external storage.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid.
ML_ERROR_STREAMS_PIPE	Failed to start the pipeline.
ML_ERROR_OUT_OF_MEMORY	Failed to allocate required memory. ml_single_h single = NULL; ml_single_open (&single, test_model, NULL, NULL, ML_NNFW_TYPE_ANY, ML_NNFW_HW_ANY); ... do something .... ml_single_close (single);

int ml_single_open_full	(	ml_single_h *	single,
		const char *	model,
		const ml_tensors_info_h	input_info,
		const ml_tensors_info_h	output_info,
		ml_nnfw_type_e	nnfw,
		ml_nnfw_hw_e	hw,
		const char *	custom_option
	)

Opens an ML model and returns the instance as a handle with custom options.

Even if the model has flexible input data dimensions, input data frames of an instance of a model should share the same dimension.

Since :: 6.5

Remarks:: http://tizen.org/privilege/mediastorage is needed if model is relevant to media storage.; http://tizen.org/privilege/externalstorage is needed if model is relevant to external storage.; The single should be released using ml_single_close().

Parameters:

[out]	single	This is the model handle opened.
[in]	model	This is the path to the neural network model file.
[in]	input_info	This is required if the given model has flexible input dimension, where the input dimension MUST be given before executing the model. It is required by some custom filters of NNStreamer. You may set NULL if it's not required.
[in]	output_info	This is required if the given model has flexible output dimension.
[in]	nnfw	The neural network framework used to open the given model. Set ML_NNFW_TYPE_ANY to let it auto-detect.
[in]	hw	Tell the corresponding nnfw to use a specific hardware. Set ML_NNFW_HW_ANY if it does not matter.
[in]	custom_option	Comma separated list of options. Use this parameter to fine-tune and optimize specific neural network framework. (e.g. NumThreads:N to set the number of threads in TensorFlow-Lite) You may set NULL if it's not required. See NNStreamer documentation (https://nnstreamer.github.io/gst/nnstreamer/tensor_filter/README.html) for the details.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_PERMISSION_DENIED	The application does not have the privilege to access to the media storage or external storage.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid.
ML_ERROR_STREAMS_PIPE	Failed to start the pipeline.
ML_ERROR_OUT_OF_MEMORY	Failed to allocate required memory. ml_single_h single = NULL; ml_single_open_full (&single, test_model, NULL, NULL, ML_NNFW_TYPE_ANY, ML_NNFW_HW_ANY, "custom_option"); ... do something .... ml_single_close (single);

int ml_single_open_with_option	(	ml_single_h *	single,
		const ml_option_h	option
	)

Makes a single instance with given ml-option.

Since :: 7.0

Remarks:: http://tizen.org/privilege/mediastorage is needed if option is relevant to media storage.; http://tizen.org/privilege/externalstorage is needed if option is relevant to external storage.; The single should be released using ml_single_close().

Parameters:

[out]	single	This is the model handle opened.
[in]	option	The handle of ml-option.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_PERMISSION_DENIED	The application does not have the privilege to access to the media storage or external storage.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid.
ML_ERROR_STREAMS_PIPE	Failed to start the pipeline.
ML_ERROR_OUT_OF_MEMORY	Failed to allocate required memory. ml_single_h single = NULL; ml_single_open_with_option (single, option); ... do something .... ml_single_close (single);

int ml_single_set_input_info	(	ml_single_h	single,
		const ml_tensors_info_h	info
	)

Sets the information (tensor dimension, type, name and so on) of required input data for the given model.

Note that a model/framework may not support changing the information. Use ml_single_get_input_info() and ml_single_get_output_info() instead for this framework.

Since :: 6.0

Parameters:

[in]	single	The model handle.
[in]	info	The handle of input tensors information.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid.

int ml_single_set_property	(	ml_single_h	single,
		const char *	name,
		const char *	value
	)

Sets the property value for the given model.

Note that a model/framework may not support changing the property after opening the model.

Since :: 6.0

Parameters:

[in]	single	The model handle.
[in]	name	The property name.
[in]	value	The property value.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid.

int ml_single_set_timeout	(	ml_single_h	single,
		unsigned int	timeout
	)

Sets the maximum amount of time to wait for an output, in milliseconds.

Since :: 5.5

Parameters:

[in]	single	The model handle.
[in]	timeout	The time to wait for an output.

Returns:: 0 on success. Otherwise a negative error value.

Return values:

ML_ERROR_NONE	Successful.
ML_ERROR_NOT_SUPPORTED	Not supported.
ML_ERROR_INVALID_PARAMETER	Fail. The parameter is invalid.

Required Header

Overview

Related Features

Functions

Typedefs

Typedef Documentation

Function Documentation