cerebras.modelzoo.common.run_utils#
Utilities for running Cerebras Pytorch Models
Functions
Checks if a given arg matches the given keyword |
|
Entry point to running pytorch models |
|
Entry point to running pytorch models including CLI argument parsing |
|
Runs a full end-to-end CS/non-CS workflow for a given model |
|
Assert the list of log levels is valid |
|
Temporary support for running eval multiple times via subprocess |
|
Temporary support for running train and eval multiple times via subprocess |
|
Starts a distributed GPU run using torchrun |
|
Updates sideband arguments to a different mode |