DatacacheConfiguration Class
Note
This is an experimental class, and may change at any time. Please see https://aka.ms/azuremlexperimental for more information.
Represents how to consume datacache in a run configuration.
This class is used in the RunConfiguration class.
Class DatacacheConfiguration constructor.
Constructor
DatacacheConfiguration(dataset_id=None, datacache_store=None, datacache_id=None, mode='mount', path_on_compute=None, replica=None, failure_fallback=False)
Parameters
| Name | Description |
|---|---|
|
datacache_id
Required
|
Id of the datacache to be consumed in the run. |
|
datacache_store
Required
|
Name of the datacache store. |
|
dataset_id
Required
|
Id of the dataset. |
|
mode
Required
|
The mode to handle. |
|
replica
Required
|
Number of replicas to be used in the job.. |
|
failure_fallback
Required
|
Whether to fallback to the original dataset if replicas are not ready. |
|
path_on_compute
Required
|
Path where data will be available on the compute. |
Methods
| create |
Create a DatacacheConfiguration from a DatacacheConsumptionConfig. |
create
Create a DatacacheConfiguration from a DatacacheConsumptionConfig.
static create(data)
Parameters
| Name | Description |
|---|---|
|
data
Required
|
the DatacacheConsumptionConfig to create the DatacacheConfiguration from. |
Returns
| Type | Description |
|---|---|
|
the DatacacheConfiguration representing how to deliver the data to the compute target. |