Skip to content

Releases: NVIDIA/MinkowskiEngine

v0.5.4

21 May 19:58
Compare
Choose a tag to compare
  • Fix TensorField.sparse() for no duplicate coordinates
  • Skip unnecessary spmm if SparseTensor.initialize_coordinates() has no duplicate coordinates
  • Model summary utility function added
  • TensorField.splat function for splat features to a sparse tensor
  • SparseTensor.interpolate function for extracting interpolated features
  • coordinate_key property function for SparseTensor and TensorField
  • Fix .dense() for GPU tensors. (PR #319)

v0.5.3

14 Apr 03:45
Compare
Choose a tag to compare
  • Updated README for pytorch 1.8.1 support
  • Use custom gpu_storage instead of thrust vector for faster constructors
  • pytorch installation instruction updates
  • fix transpose kernel map with kernel_size == stride_size
  • Update reconstruction and vae examples for v0.5 API
  • stack_unet.py example, API updates
  • MinkowskiToFeature layer

v0.5.2

05 Mar 12:59
Compare
Choose a tag to compare
  • spmm average cuda function
  • SparseTensor list operators (cat, mean, sum, var)
  • MinkowskiStack containers
  • Replace all at::cuda::getCurrentCUDASparseHandle with custom getCurrentCUDASparseHandle (issue #308)
  • fix coordinate manager kernel map python function
  • direct max pool
    • SparseTensorQuantizationMode.MAX_POOL
  • TensorField global max pool
    • origin field
    • origin field map
    • MinkowskiGlobalMaxPool CPU/GPU updates for a field input
  • SparseTensor.dense() raises a value error when a coordinate is negative rather than subtracting the minimum coordinate from a sparse tensor. (issue #316)
  • Added to_sparse() that removes zeros. (issue #317)
    • Previous to_sparse() was renamed to to_sparse_all()
    • MinkowskiToSparseTensor takes an optional remove_zeros boolean argument.
  • Fix global max pool with batch size 1
  • Use separate memory chunks for in, out map, and kernel indices for gpu_kernel_map for gpu memory misaligned error

v0.5.1

06 Feb 00:27
Compare
Choose a tag to compare
  • v0.5 documentation updates
  • Nonlinear functionals and modules
  • Warning when using cuda without ME cuda support
  • diagnostics test
  • TensorField slice
    • Cache the unique map and inverse map pair in the coordinate manager
    • generate inverse_mapping on the fly
  • CoordinateManager
    • field_to_sparse_insert_and_map
    • exists_field_to_sparse
    • get_field_to_sparse_map
    • fix kernel_map with empty coordinate maps
  • CoordiateFieldMap
    • quantize_coordinates
  • TensorField binary ops fix
  • MinkowskiSyncBatchNorm
    • Support tfield
    • convert sync batchnorm updates
  • TensorField to sparse with coordinate map key
  • Sparse matrix multiplication
    • force contiguous matrix
  • Fix AveragePooling cudaErrorMisalignedAddress error for CUDA 10 (#246)

v0.5.0

31 Dec 04:18
Compare
Choose a tag to compare

Major speedup with GPU coordinate map/manager backend.

v0.4.3

11 Jun 17:58
b683e25
Compare
Choose a tag to compare
Update terminology.rst (#156)

Correct kernel map for Kernel I. 
 - According to the animation the input with index 0 maps also to output with index 0 in kernel I.