swanlab.Settings

Settings(
    model_config = ConfigDict(frozen=True),
    metadata_collect: StrictBool = True,
    collect_hardware: StrictBool = True,
    collect_runtime: StrictBool = True,
    security_mask: StrictBool = True,
    requirements_collect: StrictBool = True,
    conda_collect: StrictBool = False,
    hardware_monitor: StrictBool = True,
    disk_io_dir: DirectoryPath = Field(...),
    upload_interval: PositiveInt = 1,
    max_log_length: int = Field(ge=500, le=4096, default=1024),
    log_proxy_type: Literal["all", "stdout", "stderr", "none"] = "all",
)

Parameter	Type	Description
metadata_collect	StrictBool	Whether to enable metadata collection. Default is True.
collect_hardware	StrictBool	Whether to collect hardware information of the current system environment. Default is True.
collect_runtime	StrictBool	Whether to collect runtime information. Default is True.
security_mask	StrictBool	Whether to automatically mask privacy information, such as api_key, etc. When enabled, any detected privacy information will be replaced with encrypted characters (****). Default is True.
requirements_collect	StrictBool	Whether to collect Python environment information (pip list). Default is True.
conda_collect	StrictBool	Whether to collect Conda environment information. Default is False.
hardware_monitor	StrictBool	Whether to enable hardware monitoring. If metadata_collect is disabled, this setting is ineffective. Default is True.
disk_io_dir	DirectoryPath	The path for disk I/O monitoring. Default is the system root directory (/ or C:\).
upload_interval	PositiveInt	Log upload interval (in seconds). Default is 1.
max_log_length	int	Maximum characters per line for terminal log upload (range: 500-4096). Default is 1024.
log_proxy_type	Literal	The log proxy type determines which content is recorded in the experiment's log tab. The default value is "all". "stdout" proxies only the standard output stream. "stderr" proxies only the standard error stream. "all" proxies both standard output and standard error streams. "none" disables log proxying.

Introduction

• The swanlab.Settings class is used to manage SwanLab's global feature switches and settings.

• When import swanlab is executed, a default global settings object is created. The settings and their default values are detailed in the table above.

• To adjust certain settings, you need to create a new Settings instance (e.g., new_settings), pass the configuration parameters you wish to modify during instantiation, and then update the global settings by running swanlab.merge_settings(new_settings).

• Note that the merge_settings() method is only available before swanlab.init() is called. This means that once swanlab.init() is invoked during the use of swanlab, the global settings can no longer be modified.

More Usage Examples

Updating Global Settings

import swanlab
from swanlab import Settings

# Create a new settings object
new_settings = Settings(
    metadata_collect=False,
    hardware_monitor=False,
    upload_interval=5
)

# Update global settings
swanlab.merge_settings(new_settings)

swanlab.init()
...

Recording Conda Environment Information

import swanlab
from swanlab import Settings

# Create a new settings object
new_settings = Settings(
    conda_collect=True  # Disabled by default
)

# Update global settings
swanlab.merge_settings(new_settings)

swanlab.init()
...