How to configure TensorFlow Lite launcher¶
TensorFlow Lite launcher is one of the supported wrappers for easily launching models within Accuracy Checker tool. This launcher allows to execute models in TensorFlow Lite framework.
For enabling TensorFlow Lite launcher you need to add framework: tf_lite
in launchers section of your configuration file and provide following parameters:
model
- path to file with TFLite model for your topology.adapter
- approach how raw output will be converted to representation of dataset problem, some adapters can be specific to framework. You can find detailed instruction how to use adapters here.device
- specifies which device will be used for infer (cpu
orgpu
).
Specifying model inputs in config.¶
In case when you model has several inputs you should provide list of input layers in launcher config section using key inputs
. Each input description should has following info:
name
- input layer name in networktype
- type of input values, it has impact on filling policy. Available options:CONST_INPUT
- input will be filled using constant provided in config. It also requires to providevalue
.IMAGE_INFO
- specific key for setting information about input shape to layer (used in Faster RCNN based topologies). You do not need to providevalue
, because it will be calculated in runtime. Format value is list withN
elements of the form[H, W, S]
, whereN
is batch size,H
- original image height,W
- original image width,S
- scale of original image (default 1).ORIG_IMAGE_INFO
- specific key for setting information about original image size before preprocessing.PROCESSED_IMAGE_INFO
- specific key for setting information about input size after preprocessing.SCALE_FACTOR
- specific key for setting information about image scale factor defined as[SCALE_Y, SCALE_X]
, whereSCALE_Y
=<resized_image_height>/<original_image_height
,SCALE_X
=<resized_image_width> / <original_image_width>
IGNORE_INPUT
- input which should be stayed empty during evaluation.INPUT
- network input for main data stream (e. g. images). If you have several data inputs, you should provide regular expression for identifier asvalue
for specifying which one data should be provided in specific input.
Optionally you can determine
shape
of input andlayout
in case when your model was trained with non-standard data layout (For TensorFlow Lite default layout isNHWC
) andprecision
(Supported precisions:FP32
- float,FP16
- signed shot,U8
- unsigned char,U16
- unsigned short int,I8
- signed char,I16
- short int,I32
- int,I64
- long int).
TensorFlow Lite launcher config example:
launchers:
- framework: tf_lite
device: CPU
model: path_to_model/alexnet.tflite
adapter: classification