mirror of
https://github.com/wassname/Ranger21.git
synced 2026-09-10 11:40:27 +08:00
Update README.md
This commit is contained in:
@@ -33,7 +33,7 @@ In addition, training a BN Free (no batch norm) ResNet50 as an additional compar
|
||||

|
||||
|
||||
|
||||
* Ranger21 code updates - due to firsthand experience, have added in safety guards in the event that num_epochs does not match the actual epochs trained, as well as updated the linear warmdown code to be simpler and never go below the min_lr designated (defaults to 3e-5). This will start to spew a lot of text to alert you on each iteration, but the lr will be automatically guarded. </br>
|
||||
* Ranger21 code updates - due to firsthand experience, have added in safety guards in the event that num_epochs set for Ranger21 does not match the actual epochs being run, as well as updated the linear warmdown code to be simpler and never go below the min_lr designated (defaults to 3e-5). </br>If there is an epoch mis-match between num_epochs passed to optimizer and the atual run, this will start to spew a lot of text to alert you on each iteration, but the lr itself will now be automatically guarded and not go below the min_lr. </br>
|
||||
 </br>
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user