423A35C7/pytorch3d
1
0
Fork 0
mirror of https://github.com/facebookresearch/pytorch3d.git synced 2025-12-20 06:10:34 +08:00
Code Issues Packages Projects Releases Wiki Activity

Better seeding of random engines

Browse Source 
Summary: Currently, seeds are set only inside the train loop. But this does not ensure that the model weights are initialized the same way everywhere which makes all experiments irreproducible. This diff fixes it.

Reviewed By: bottler

Differential Revision: D38315840

fbshipit-source-id: 3d2ecebbc36072c2b68dd3cd8c5e30708e7dd808
This commit is contained in:
David Novotny
2022-08-01 10:03:09 -07:00
committed by Facebook GitHub Bot
parent 0c3599e8ee
commit 80fc0ee0b6
4 changed files with 31 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
projects/implicitron_trainer/tests/experiment.yaml
View File

@@ -2,6 +2,7 @@ data_source_class_type: ImplicitronDataSource
model_factory_class_type: ImplicitronModelFactory
optimizer_factory_class_type: ImplicitronOptimizerFactory
training_loop_class_type: ImplicitronTrainingLoop
seed: 42
detect_anomaly: false
exp_dir: ./data/default_experiment/
hydra:
@@ -429,7 +430,6 @@ training_loop_ImplicitronTrainingLoop_args:
eval_only: false
evaluator_class_type: ImplicitronEvaluator
max_epochs: 1000
seed: 0
store_checkpoints: true
store_checkpoints_purge: 1
test_interval: -1