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

more padding options in Dataloader

Browse Source 
Summary: Add facilities for dataloading non-sequential scenes.

Reviewed By: shapovalov

Differential Revision: D37291277

fbshipit-source-id: 0a33e3727b44c4f0cba3a2abe9b12f40d2a20447
This commit is contained in:
Jeremy Reizenstein
2022-07-06 07:13:41 -07:00
committed by Facebook GitHub Bot
parent 0dce883241
commit 771cf8a328
16 changed files with 449 additions and 70 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
projects/implicitron_trainer/tests/experiment.yaml
View File

@@ -345,10 +345,13 @@ data_source_args:
data_loader_map_provider_SequenceDataLoaderMapProvider_args:
batch_size: 1
num_workers: 0
dataset_len: 1000
dataset_len_val: 1
images_per_seq_options:
- 2
dataset_length_train: 0
dataset_length_val: 0
dataset_length_test: 0
train_conditioning_type: SAME
val_conditioning_type: SAME
test_conditioning_type: KNOWN
images_per_seq_options: []
sample_consecutive_frames: false
consecutive_frames_max_gap: 0
consecutive_frames_max_gap_seconds: 0.1

2
projects/implicitron_trainer/tests/test_experiment.py
View File

@@ -55,7 +55,7 @@ class TestExperiment(unittest.TestCase):
dataset_args.test_restrict_sequence_id = 0
dataset_args.dataset_root = "manifold://co3d/tree/extracted"
dataset_args.dataset_JsonIndexDataset_args.limit_sequences_to = 5
dataloader_args.dataset_len = 1
dataloader_args.dataset_length_train = 1
cfg.solver_args.max_epochs = 2
device = torch.device("cuda:0")