andrea.pedrotti
/
gFun
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Implemented custom micro and macro F1 in pl (cpu and gpu)

This commit is contained in:
andrea 2021-01-20 14:56:06 +01:00
parent a60e2cfc09
commit faa387f696
1 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
refactor/models/pl_gru.py
View File

@ -69,8 +69,9 @@ class RecurrentModel(pl.LightningModule):
self.linear2 = nn.Linear(ff1, ff2)
self.label = nn.Linear(ff2, self.output_size)
lPretrained = None # TODO: setting lPretrained to None, letting it to its original value will bug first
# validation step (i.e., checkpoint will store also its ++ value, I guess, making the saving process too slow)
# TODO: setting lPretrained to None, letting it to its original value will bug first validation
# step (i.e., checkpoint will store also its ++ value, I guess, making the saving process too slow)
lPretrained = None
self.save_hyperparameters()
def forward(self, lX):