... | @@ -40,6 +40,15 @@ To begin with, we will put a taskwait between each call of the main training loo |
... | @@ -40,6 +40,15 @@ To begin with, we will put a taskwait between each call of the main training loo |
|
|
|
|
|

|
|

|
|
|
|
|
|
|
|
Which gives us these paraver traces :
|
|
|
|

|
|
|
|
|
|
|
|

|
|
|
|
|
|
|
|

|
|
|
|
|
|
|
|

|
|
|
|
|
|
This lead us to the following performances :
|
|
This lead us to the following performances :
|
|
|
|
|
|
Here we tried to slice according to the token sequence only (BATCH_SUBSIZE=4), or according to the sequence token and the sequence batch (BATCH_SUBSIZE=1). The slope obtained for the second version seems better so we will keep it for the following tests.
|
|
Here we tried to slice according to the token sequence only (BATCH_SUBSIZE=4), or according to the sequence token and the sequence batch (BATCH_SUBSIZE=1). The slope obtained for the second version seems better so we will keep it for the following tests.
|
... | | ... | |