aclInit

configPath

Input

Pointer to the path (including the file name) of the configuration file. The configuration file is in .json format. A .json file allows up to 10 levels of curly brackets and square brackets, respectively.

When initializing AscendCL, you can use this configuration file to enable or set the following functions. If the following default configurations meet the requirements, you do not need to modify them. You can pass NULL to aclInit or set the configuration file to an empty JSON string (that is, the configuration file contains only {}).

• Dump information configuration, including:
  • Model dump configuration (used to export the input and output data of operators at each layer in the model) and single-operator dump configuration (used to export the input and output data of an operator). The exported data is used to compare with that of a specified model or operator to locate accuracy issues. For details about the configuration example, description, and restrictions, see Configuration File Example (Model Dump and Single-Operator Dump). Dump is disabled by default.

    If you use this API to enable the dump configuration, you need to set the dump_path parameter to configure the path where the dump data is flushed to drives. If the dump data does not need to be flushed to drives, you can use the callback function to obtain the dump data. For details, see acldumpRegCallback.

  • Dump configuration of the exception operator (used to export the input and output data, workspace information, and tiling information of the exception operator). The exported data is used to analyze AI Core errors. For details about the configuration example, see Configuration File Example (Exception Operator Dump Configuration). Dump is disabled by default.
  • Overflow/Underflow operator dump configuration (used to export the input and output data of overflow/underflow operators in the model). The exported data is used to analyze overflow/underflow causes and locate model accuracy issues. For details about the configuration example, description, and restrictions, see Configuration File Example (Overflow/Underflow Operator Dump Configuration). Dump is disabled by default.
  • Configuration for operator dump watch mode (used to enable the observation mode for the output data of a specified operator). If you suspect that the memory is overwritten by other operators after locating the accuracy issues of some operators and excluding the calculation issues of the operators, you can enable the dump watch mode. For details about the configuration example and restrictions, see Configuration File Example (Configuration for Operator Dump Watch Mode). The dump watch mode is disabled by default.
• Profiling configuration: For details about the configuration example, description, and restrictions, see Profile Data Collection with the acl.json Configuration File in Performance Tuning Tool User Guide . Profiling is disabled by default.
• Operator cache aging configuration. To reduce the memory footprint and balance call performance when you execute a single operator in the single-operator model execution mode (except when you use aclopUpdateParams), you can use the max_opqueue_num parameter to configure the maximum length of the "operator type and single-operator model" mapping queue. If the length of the mapping queue reaches the maximum, the least used mapping information and single-operator models in the cache are deleted before the most recent mapping information and the corresponding single-operator model are loaded. The default maximum length of the mapping queue is 20000. For the configuration example and restrictions, see Configuration File Example (Operator Cache Aging Configuration).

  For details about the single-operator model, see Single-Operator Call Sequence.

• Configuration for error information report mode, which is used to control the aclGetRecentErrMsg API to obtain error information by process or thread level. By default, error information is obtained by thread level. For details about the example, see Configuration File Example (Configuration for Error Information Reporting Mode).
• Default device configuration (used to configure the default compute device). For details about the configuration example and description, see Example (Default Device Configuration).

  If the device is specified through aclrtSetDevice, aclrtSetDevice has a high priority.

  If you want to explicitly create a context after enabling the default device configuration, you need to call aclrtSetDevice. Otherwise, service exceptions may occur.

dump_list

(Required) List of network-wide models for data dump.

Create model dump configuration information. If multiple models need to be dumped, separate them with commas (,).

In the single-operator calling scenario (including single-operator model execution and single-operator API execution), dump_list is empty.

model_name

Model name. The value of model_name of each model must be unique.

• To load a model from a file, enter the model file name without the name extension. You can also set this parameter to the value of the outermost name field in the .json file after ATC-based model conversion.
• To load a model from memory, set this parameter to the value of the name field in the .json file after ATC-based model conversion.

layer

It is advised to dump certain operators only. Otherwise, excessive data may induce timeouts if the I/O performance is poor. This field can be used to specify the name of the operator to be dumped. The name can be the name of the operator after ATC model conversion or the name of the original operator before conversion.

• Configure the operator name in each line in the format. Use commas (,) to separate operators.
• You do not need to set model_name. In this case, the corresponding operators of all models are dumped by default. If model_name is set, the corresponding operators of the model are dumped.
• If the input of the specified operator involves the data operator, the data operator information is dumped. To dump the data operator, enter the downstream nodes of the data operator.
• To dump all operators of a model, the layer field does not need to be included.

dump_path

(Required) Directory for storing dump data files in the operating environment. The directory must be created in advance and the running user configured during installation must have the read and write permissions on the directory.

dump_mode

Dump mode.

• input: dumps operator inputs only.
• output (default): dumps operator outputs only.
• all: dumps both operator inputs and outputs.

  Note: If this parameter is set to all, the input data of some operators, such as collective communication operators HcomAllGather and HcomAllReduce, will be modified during execution. Therefore, the system dumps the operator input before operator execution and dumps the operator output after operator execution. In this way, the dumped input and output data of the same operator is flushed to drives separately, and multiple dump files are generated. After parsing the dump files, you can determine whether the data is an input or output based on the file content.

dump_level

Dump data level. The options are as follows:

• op: dumps data at the operator level.
• kernel: dumps data at the kernel level.
• all (default): dumps both op and kernel level data.

If the default value is used, there are a large number of dump files, for example, dump files starting with aclnn. If you have requirements on the dump performance or the memory resources are limited, you can set this parameter to the op level to improve the dump performance and reduce the number of dump files.

dump_op_switch

Dump data switch of the single-operator model execution mode in the single-operator dump scenario.

• on: enables dump for the single-operator model.
• off (default): disables dump for the single-operator model.

dump_step

Iterations to dump. This parameter is not required in the inference scenario.

If this parameter is not configured, dump data will be generated for all iterations by default, which may result in a large amount of data. You are advised to specify iterations as required.

Separate multiple iterations using vertical bars (|), for example, 0|5|10. You can also use hyphens (-) to specify the iteration range, for example, 0|3-5|10.

Configuration example:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11

{
	"dump":{
		"dump_list":[     
			...... 
		],  
		"dump_path":"$HOME/output",
                "dump_mode":"output",
		"dump_op_switch":"off",
                "dump_step": "0|3-5|10"
	}  
}

dump_data

Type of the operator dump content. The options are as follows:

• tensor (default): dumps operator data.
• stats: dumps operator statistics. The result file is in .csv format and contains the operator name, input/output data type, maximum value, and minimum value.

Dumping a large amount of data typically requires a significant amount of time. One solution is to first dump operator statistics, use the statistics to identify potentially abnormal operators, and then proceed to dump the data of the identified operators.

In the model dump scenario, the information of operator input or output or both can be collected based on the configuration of dump_mode.