morpheus.config.Config
- class Config(execution_mode=ExecutionMode.GPU, debug=False, log_level=30, log_config_file=None, plugins=None, mode=PipelineModes.OTHER, _pipeline_batch_size=256, _model_max_batch_size=8, feature_length=256, num_threads=1, edge_buffer_size=128, class_labels=<factory>, ae=None, fil=None, frozen=False)[source]
Bases:
morpheus.config.ConfigBasePipeline configuration class.
- Parameters
- debugbool
Flag to run pipeline in debug mode. Default value is False.
- log_levelint
Specifies the log level and above to output. Must be one of the available levels in the
loggingmodule.- mode
PipelineModes Determines type of pipeline Ex: FIL or NLP. Use
OTHERfor custom pipelines.- feature_lengthint
Specifies the dimension of the second axis for messages in the pipeline. For NLP this is the sequence length. For FIL this is the number of input features.
- pipeline_batch_sizeint
Determines number of messages per batch. Default value is 256.
- num_threadsint
Number threads to process each batch of messages. Default value is 1.
- model_max_batch_size8
In a single batch, the maximum number of messages to send to the model for inference. Default value is 8.
- edge_buffer_sizeint, default = 128
The size of buffered channels to use between nodes in a pipeline. Larger values reduce backpressure at the cost of memory. Smaller values will push messages through the pipeline quicker. Must be greater than 1 and a power of 2 (i.e., 2, 4, 8, 16, etc.).
- Attributes
- ae
ConfigAutoEncoder Config for autoencoder.
- log_config_filestr
File corresponding to this Config.
- ae
Methods
- freeze()[source]
Freeze the Config object, making it immutable. This method will be invoked when the config object is passed to a pipeline or stage for the first time.
Calling
freezeon a frozen instance will not have any effect.
- save(filename)[source]
Save Config to file.
- Parameters
- filenamestr
File path to save Config.
- to_string()[source]
Get string representation of Config.
- Returns
- str
Config as string.