TensorSpace Converter - Converter API

TensorSpace Converter API

Sample TensorSpace-Converter script:

$ tensorspacejs_converter \
    --input_model_from="XXX" \
    --input_model_format="YYY" \
    --output_layer_names="EEE1,EEE2,EEE3" \
    input_path \
    output_path

Arguments explanation:

Positional Arguments	Description
input_path	Path for model input artifacts.
output_path	Folder for all output artifacts.

Options	Description
input_model_from	Configure the training library for pre-trained model, use: tensorflow for TensorFlow, keras for Keras, tfjs for TensorFlow.js.
input_model_format	The format of input model, set different attributes for different kinds of models.
output_layer_names	The names of the layer which will be visualized in TensorSpace, separated by comma ",".

The TensorSpace-Converter Example Usage Page introduce how to set Positional Arguments and Optional Arguments for different kinds of models.