kp.inference

---

kp.inference.generic_inference_retrieve_fixed_node(node_idx, generic_raw_result, channels_ordering)

Retrieve single node output fixed-point data from GenericRawResult/GenericRawBypassPreProcResult object.

• Parameters

  node_idx : int

  Wanted output node index, starts from 0. Number of total output nodes can be known
  from ‘GenericRawResult.header.num_output_node’
  

  generic_raw_result : kp.GenericRawResult, kp.GenericRawBypassPreProcResult

  GenericRawResult/GenericRawBypassPreProcResult object contained the received RAW data results, it should
  come from ‘kp.inference.generic_raw_inference_receive()’/’kp.inference.generic_raw_inference_bypass_pre_proc_receive()’
  

  channels_ordering : kp.ChannelOrdering

  The raw output feature map channel ordering.
  

• Returns

  inference_fixed_node_output : kp.InferenceFixedNodeOutput

  Raw node output in fixed-point format.
  

• Return type

  InferenceFixedNodeOutput

---

kp.inference.generic_inference_retrieve_float_node(node_idx, generic_raw_result, channels_ordering)

Retrieve single node output floating-point data from GenericRawResult/GenericRawBypassPreProcResult object.

• Parameters

  node_idx : int

  Wanted output node index, starts from 0. Number of total output nodes can be known
  from ‘GenericRawResult.header.num_output_node’
  

  generic_raw_result : kp.GenericRawResult, kp.GenericRawBypassPreProcResult

  GenericRawResult/GenericRawBypassPreProcResult object contained the received RAW data results, it should
  come from ‘kp.inference.generic_raw_inference_receive()’/’kp.inference.generic_raw_inference_bypass_pre_proc_receive()’
  

  channels_ordering : kp.ChannelOrdering

  The raw output feature map channel ordering.
  

• Returns

  inference_float_node_output : kp.InferenceFloatNodeOutput

  Raw node output in floating-point format.
  

• Return type

  InferenceFloatNodeOutput

---

kp.inference.generic_raw_inference_bypass_pre_proc_receive(device_group, generic_raw_image_header, model_nef_descriptor)

Generic raw inference bypass pre-processing receive. When an image inference is done, this function can be used to get the results in RAW format. Note that data received is in Kneron RAW format, users need kp.inference.generic_inference_retrieve_float_node()/kp.inference.generic_inference_retrieve_fixed_node() to convert RAW format data to floating-point/fixed-point data.

• Parameters

  device_group : kp.DeviceGroup

  Represents a set of devices handle.
  

  generic_raw_image_header : kp.GenericRawBypassPreProcImageHeader

  Needed parameters for performing bypass pre-processing inference including image buffer size, model ID ..etc.
  

  model_nef_descriptor : kp.ModelNefDescriptor

  ModelNefDescriptor object for describing the uploaded models.
  

• Returns

  generic_raw_result : kp.GenericRawBypassPreProcResult

  GenericRawBypassPreProcResult object contained the received RAW data results.
  

• Raises

  kp.ApiKPException

• Notes
  The data received is in Kneron RAW format, users need kp.inference.generic_inference_retrieve_float_node()/kp.inference.generic_inference_retrieve_fixed_node() to convert RAW format data to floating-point/fixed-point data.

• Return type

  GenericRawBypassPreProcResult

---

kp.inference.generic_raw_inference_bypass_pre_proc_send(device_group, generic_raw_image_header, image_buffer)

Generic raw inference bypass pre-processing send. This is to perform a single image inference, it is non-blocking if device buffer queue is not full. When this is performed, user can issue kp.inference.generic_raw_inference_bypass_pre_proc_receive() to get the result. In addition, to have better performance, users can issue multiple kp.inference.generic_raw_inference_bypass_pre_proc_send() then start to receive results through kp.inference.generic_raw_inference_bypass_pre_proc_receive().

• Parameters

  device_group : kp.DeviceGroup

  Represents a set of devices handle.
  

  generic_raw_image_header : kp.GenericRawBypassPreProcImageHeader

  Needed parameters for performing bypass pre-processing inference including image buffer size, model ID ..etc.
  

  image_buffer : bytes

  The data bytes contains the image.
  

• Raises

  kp.ApiKPException

• Return type

  None

---

kp.inference.generic_raw_inference_receive(device_group, generic_raw_image_header, model_nef_descriptor)

Generic raw inference receive. When an image inference is done, this function can be used to get the results in RAW format. Note that the data received is in Kneron RAW format, users need kp.inference.generic_inference_retrieve_float_node() to convert RAW format data to floating-point data.

• Parameters

  device_group : kp.DeviceGroup

  Represents a set of devices handle.
  

  generic_raw_image_header : kp.GenericRawImageHeader

  Needed parameters for performing inference including image width, height ..etc.
  

  model_nef_descriptor : kp.ModelNefDescriptor

  ModelNefDescriptor object for describing the uploaded models.
  

• Returns

  generic_raw_result : kp.GenericRawResult

  GenericRawResult object contained the received RAW data results.
  

• Raises

  kp.ApiKPException

• Notes
  The data received is in Kneron RAW format, users need kp.inference.generic_inference_retrieve_float_node() to convert RAW format data to floating-point data.

• Return type

  GenericRawResult

---

kp.inference.generic_raw_inference_send(device_group, generic_raw_image_header, image, image_format)

Generic raw inference send. This is to perform a single image inference, it is non-blocking if device buffer queue is not full. When this is performed, user can issue kp.inference.generic_raw_inference_receive() to get the result. In addition, to have better performance, users can issue multiple kp.inference.generic_raw_inference_send() then start to receive results through kp.inference.generic_raw_inference_receive().

• Parameters

  device_group : kp.DeviceGroup

  Represents a set of devices handle.
  

  generic_raw_image_header : kp.GenericRawImageHeader

  Needed parameters for performing inference including image width, height ..etc.
  

  image : bytes, numpy.ndarray

  The data bytes or numpy.ndarray contains the image.
  

  image_format : kp.ImageFormat

  Image format supported for inference.
  

• Raises

  kp.ApiKPException

• Return type

  None

---

kp.inference.set_inference_configuration(device_group, inference_configuration)

Configure inference settings.

• Parameters

  device_group : kp.DeviceGroup

  Represents a set of devices handle.
  

  inference_configuration : kp.InferenceConfiguration

  Inference configurations.
  

• Raises

  kp.ApiKPException

• Return type

  None