Actions: pyg-team/pytorch_geometric
Actions
Showing runs from all workflows
35,930 workflow run results
35,930 workflow run results
InLUT3D
point cloud dataset
Documentation
#19228:
Pull request #9474
synchronize
by
jamesWalczak
InLUT3D
point cloud dataset
Testing previous PyTorch versions
#8966:
Pull request #9474
synchronize
by
jamesWalczak
InLUT3D
point cloud dataset
Linting
#19128:
Pull request #9474
synchronize
by
jamesWalczak
InLUT3D
point cloud dataset
Testing minimal PyTorch 2.5
#6967:
Pull request #9474
synchronize
by
jamesWalczak
InLUT3D
point cloud dataset
Testing PyTorch 2.5
#19395:
Pull request #9474
synchronize
by
jamesWalczak
InLUT3D
point cloud dataset
PR Labeler
#12907:
Pull request #9474
synchronize
by
jamesWalczak
InLUT3D
point cloud dataset
Testing distributed PyG
#2092:
Pull request #9474
synchronize
by
jamesWalczak
InLUT3D
point cloud dataset
Changelog Enforcer
#20394:
Pull request #9474
synchronize
by
jamesWalczak
visualize_feature_importance
in heterogenous graphs with single features
Changelog Enforcer
#20393:
Pull request #9499
opened
by
nilserranestle
visualize_feature_importance
in heterogenous graphs with single features
PR Labeler
#12906:
Pull request #9499
opened
by
nilserranestle
visualize_feature_importance
in heterogenous graphs with single features
Testing PyTorch nightly
#8549:
Pull request #9499
opened
by
nilserranestle
visualize_feature_importance
in heterogenous graphs with single features
Documentation
#19227:
Pull request #9499
opened
by
nilserranestle
visualize_feature_importance
in heterogenous graphs with single features
Testing minimal PyTorch 2.5
#6966:
Pull request #9499
opened
by
nilserranestle
visualize_feature_importance
in heterogenous graphs with single features
Testing previous PyTorch versions
#8965:
Pull request #9499
opened
by
nilserranestle
visualize_feature_importance
in heterogenous graphs with single features
Testing distributed PyG
#2091:
Pull request #9499
opened
by
nilserranestle
visualize_feature_importance
in heterogenous graphs with single features
Testing PyTorch 2.5
#19394:
Pull request #9499
opened
by
nilserranestle
visualize_feature_importance
in heterogenous graphs with single features
Linting
#19127:
Pull request #9499
opened
by
nilserranestle
str(data)
correctly
PR Labeler
#12905:
Pull request #9498
synchronize
by
hnsgrvr
str(data)
correctly
Changelog Enforcer
#20392:
Pull request #9498
synchronize
by
hnsgrvr
str(data)
correctly
Testing distributed PyG
#2090:
Pull request #9498
synchronize
by
hnsgrvr
str(data)
correctly
Testing minimal PyTorch 2.5
#6965:
Pull request #9498
synchronize
by
hnsgrvr
str(data)
correctly
Testing PyTorch 2.5
#19393:
Pull request #9498
synchronize
by
hnsgrvr