a class for parameter servers More...
Inheritance diagram for nabu.neuralnetworks.trainers.trainer.ParameterServer:
Public Member Functions | |
| def | __init__ (self, conf, modelconf, dataconf, server, task_index) |
| NnetTrainer constructor, creates the training graph. More... | |
| def | join (self) |
| wait for the workers to finish | |
Public Attributes | |
| graph | |
| server | |
| task_index | |
| wait_op | |
| scaffold | |
Detailed Description
a class for parameter servers
Constructor & Destructor Documentation
§ __init__()
| def nabu.neuralnetworks.trainers.trainer.ParameterServer.__init__ | ( | self, | |
| conf, | |||
| modelconf, | |||
| dataconf, | |||
| server, | |||
| task_index | |||
| ) |
NnetTrainer constructor, creates the training graph.
- Parameters
-
conf the trainer config modelconf the model configuration dataconf the data configuration as a ConfigParser server optional server to be used for distributed training task_index optional index of the worker task in the cluster
The documentation for this class was generated from the following file:
