forked from mindspore-Ecosystem/mindspore
!5063 modify sgd and momentum and WithGradCell comments
Merge pull request !5063 from lijiaqi/momentum_and_sgd
This commit is contained in:
commit
15ae3702f9
|
@ -82,7 +82,7 @@ class WithGradCell(Cell):
|
||||||
|
|
||||||
Wraps the network with backward cell to compute gradients. A network with a loss function is necessary
|
Wraps the network with backward cell to compute gradients. A network with a loss function is necessary
|
||||||
as argument. If loss function in None, the network must be a wrapper of network and loss function. This
|
as argument. If loss function in None, the network must be a wrapper of network and loss function. This
|
||||||
Cell accepts *inputs as inputs and returns gradients for each trainable parameter.
|
Cell accepts '*inputs' as inputs and returns gradients for each trainable parameter.
|
||||||
|
|
||||||
Note:
|
Note:
|
||||||
Run in PyNative mode.
|
Run in PyNative mode.
|
||||||
|
|
Loading…
Reference in New Issue