For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).
Python module
max.pipelines.logging_utils
Pipeline logging utilities.
Free functions for logging pipeline and architecture information at startup.
These live outside PipelineConfig so that
config.py stays free of module-level registry imports — config is pure
data; the registry provides the arch context needed for display.
log_basic_config()
max.pipelines.logging_utils.log_basic_config(pipeline_config)
Logs minimal pipeline configuration information.
Logs basic pipeline options including model name, pipeline task, weight path, max_batch_size, max_seq_len, and reserved memory.
-
Parameters:
-
pipeline_config (PipelineConfig) – The resolved pipeline configuration to log.
-
Raises:
-
ValueError – If no architecture is found for the model. This should not happen after config resolution.
-
Return type:
-
None
log_pipeline_info()
max.pipelines.logging_utils.log_pipeline_info(pipeline_config)
Logs comprehensive pipeline and KVCache configuration information.
-
Parameters:
-
pipeline_config (PipelineConfig) – The resolved pipeline configuration to log.
-
Raises:
-
ValueError – If no architecture is found for the model. This should not happen after config resolution.
-
Return type:
-
None
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!