423A35C7/pytorch3d
1
0
Fork 0
mirror of https://github.com/facebookresearch/pytorch3d.git synced 2025-12-20 22:30:35 +08:00
Code Issues Packages Projects Releases Wiki Activity

CO3Dv2 trainer configs

Browse Source 
Summary:
Adds yaml configs to train selected methods on CO3Dv2.

Few more updates:
1) moved some fields to base classes so that we can check is_multisequence in experiment.py
2) skip loading all train cameras for multisequence datasets, without this, co3d-fewview is untrainable
3) fix bug in json index dataset provider v2

Reviewed By: kjchalup

Differential Revision: D38952755

fbshipit-source-id: 3edac6fc8e20775aa70400bd73a0e6d52b091e0c
This commit is contained in:
David Novotny
2022-08-30 13:42:19 -07:00
committed by Facebook GitHub Bot
parent 03562d87f5
commit 1163eaab43
17 changed files with 120 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
projects/implicitron_trainer/tests/experiment.yaml
View File

@@ -406,8 +406,13 @@ optimizer_factory_ImplicitronOptimizerFactory_args:
linear_exponential_lr_milestone: 200
linear_exponential_start_gamma: 0.1
training_loop_ImplicitronTrainingLoop_args:
eval_only: false
evaluator_class_type: ImplicitronEvaluator
evaluator_ImplicitronEvaluator_args:
is_multisequence: false
camera_difficulty_bin_breaks:
- 0.97
- 0.98
eval_only: false
max_epochs: 1000
store_checkpoints: true
store_checkpoints_purge: 1
@@ -420,8 +425,3 @@ training_loop_ImplicitronTrainingLoop_args:
visdom_env: ''
visdom_port: 8097
visdom_server: http://127.0.0.1
evaluator_ImplicitronEvaluator_args:
camera_difficulty_bin_breaks:
- 0.97
- 0.98
is_multisequence: false