Tuning Options

input_format

Sets the model input format.

input_shape

Sets the shape of each input.

op_name_map

Sets the directory (including the file name) of the mapping configuration file of a custom operator. The function of a custom operator varies according to the network. You can specify the mapping between the custom operator and the actual custom operator running on the network.

dynamic_batch_size

Sets dynamic batch size profiles. Applies to the scenario where image count per inference batch is unfixed.

dynamic_image_size

Sets dynamic image size profiles. Applies to the scenario where the resolution of images input for inference is unfixed.

dynamic_dims

Sets dynamic dimension profiles in ND format. Applies to the scenario where the dimensions for inference are unfixed.

precision_mode

Selects the operator precision mode.

output_type

Sets the output data type of a network or an output node.

out_nodes

Sets the output nodes.

input_fp16_nodes

Sets the name of the FP16 input node.

log

Sets the log level during tuning.

op_debug_level

Enables TBE operator debug during operator compilation.

insert_op_conf

Sets the configuration file directory (including the file name) of an operator to be inserted, for example, the AIPP operator for data preprocessing.

After this option is used, the input data type must be UINT8.

output

Sets the path of the tuned model, including the file name.

reload

Reloads tuning after subgraph tuning is interrupted. After the current process is interrupted, if you want to continue tuning from the previous phase, run --reload to enter the reload mode.

framework

Sets the framework of the original network model.

model_path

Sets the path for storing one or more original model files.

tune_ops_file

Specifies the operator name or operator type in the configuration file to tune a specified operator.

op_precision_mode

Sets the precision mode of an operator. You can use this option to set different precision modes for different operators.

keep_dtype

Keeps the computation precision of some operators unchanged during the compilation of the original network model.

singleop

Tunes one or more specified operators by configuring the operator description file.

tune_optimization_level

Sets the tuning mode, including the high-performance mode and normal mode.

Fdeeper_opat

Sets in-depth operator tuning.

Fnonhomo_split

Sets non-uniform subgraph partition tuning.

Fop_format

Sets operator format tuning.

host_env_os

If the OS and its architecture of the model compilation environment are inconsistent with those of the model operating environment, set this option to the OS type of the model operating environment. If this option is not set, the OS type of the model compilation environment is used by default.

host_env_cpu

If the OS and its architecture of the model compilation environment are inconsistent with those of the model operating environment, set this option to the OS architecture of the model operating environment.

If this option is not set, the OS architecture of the model compilation environment is used by default.

modify_mixlist

Sets the operators on the mixed precision list.

build_bypass

Transparently transmits compilation options that are not detectable by the AOE tuning framework and tuning services in the model compilation phase. For details about the specific options to be transmitted, see aclgrphBuildModel.