v1.1.2

Performance and caching improvements around the union and intersection basis routines

v1.1.1

• Minor performance improvements

v1.1.0

Changes

• Elementwise operations between different tensors are now allowed, as long as both tensors have the same basis along each dimension
• Adds the attribute tensor.numpy_compatible. If True (the default), the tensor can be used in most cases like a regular NumPy array,
  however without consideration of the basis of the tensor.

v1.0.1

What's Changed

• Renames tensor.shape to tensor.current_shape