Mirrors/the-algorithm
2
0
Fork 0
mirror of https://github.com/twitter/the-algorithm.git synced 2025-01-07 01:48:16 +01:00
Code Issues Packages Projects Releases Wiki Activity

Compare commits

base: Mirrors:0ec0c2bc29391981c575c4ad2cf9819f7da8098b
Branches Tags
Mirrors:main
...
compare: Mirrors:f2184291593fe414d7e2f97952a925080f14f437
Branches Tags
Mirrors:main

3 Commits
0ec0c2bc29 ... f218429159

Author SHA1 Message Date
Harshil
f218429159
Merge 3051db2629 into fb54d8b549 2023-05-22 17:38:15 -05:00
Harshil
3051db2629
Update trust_and_safety_models/toxicity/train.py 2023-04-02 00:56:46 +05:30
Harshil Mehta
c94f4a42b9
make it easier to do multiple fold training 2023-04-01 23:58:12 +05:30
1 changed files with 13 additions and 12 deletions
Show Stats Expand all files Collapse all files

3
trust_and_safety_models/toxicity/train.py
View File

@ -387,7 +387,8 @@ class Trainer(object):
fold=i,
)
else:
raise ValueError("Sure you want to do multiple fold training")
a = input("Are you sure you want to do multiple fold training? (y/N)")
if a.lower() == "y":
for mb_generator, steps_per_epoch, val_data, test_data in self.mb_loader(full_df=df):
self._train_single_fold(
mb_generator=mb_generator,