vai_p_tensorflow Usage - 1.2 English

Vitis AI Optimizer User Guide (UG1333)

Document ID: UG1333
Release Date: 2020-07-07
Version: 1.2 English

The following arguments are available when running vai_p_tensorflow:

Table 1. vai_p_tensorflow Arguments
Argument	Type	Default	Description
workspace	string	""	Directory for saving output files.
action	string	""	Which action to run.
input_graph	string	""	Path of a GraphDef protobuf file that defines the network’s architecture.
input_ckpt	string	""	Path of a checkpoint file. It is the prefix of filenames created for the checkpoint.
output_graph	string	""	Path of a GraphDef protobuf file for saving the pruned network.
output_ckpt	string	""	Path of a checkpoint file for saving weights.
eval_fn_path	string	""	Path of a file defining model_fn function.
max_num_batches	int	None	Max number of batches to evaluate by default use all.
target	string	""	The output node name that indicates the model’s performance.
gpu	string	""	GPU device IDs to use separated by ‘,’.
sparsity	float	None	The desired sparsity of network after pruning.
exclude	repeated	None	Convolution nodes excluded from pruning.
input_nodes	repeated	None	Input nodes of the inference graph.
input_node_shapes	repeated	None	Shape of input nodes.
output_nodes	repeated	None	Output nodes of the inference graph.
channel_batch	int	2	The number of output channels is a multiple of this value after pruning.