Searched refs:sparse (Results 1 – 8 of 8) sorted by relevance
127 SphericalVector.Annotated sparse = vec.toAnnotated(); in annotated() local129 assertClose(sparse, vec); in annotated()130 assertThat(sparse.isComplete()).isTrue(); in annotated()131 assertEquals("[⦡ 10.0 100%,⦨ 15.0 100%,⤠20.00 100%]", sparse.toString()); in annotated()133 sparse = vec.toAnnotated(true, false, false); in annotated()134 assertThat(sparse.hasAzimuth).isTrue(); in annotated()135 assertThat(sparse.hasElevation).isFalse(); in annotated()136 assertThat(sparse.hasDistance).isFalse(); in annotated()137 assertEquals("[⦡ 10.0 100%,⦨ x ,⤠ x ]", sparse.toString()); in annotated()139 sparse = vec.toAnnotated(false, true, false); in annotated()[all …]
49 * Expands a representation of a sparse tensor to a dense tensor.59 * * 3: In the traversal order defined above, the format (dense vs. sparse) and index metadata61 * a sparse dimension, it's the same as the compressed index defined in the Compressed87 * * 0: A 1-D tensor representing the compressed sparse tensor data of a conceptual91 * block sparse with a k-dimensional block (0 < k <= n), the traversal order has n+k94 * internally. If not block sparse, the traversal order is just a permutation of [D0, …,97 * block sparse n-dimensional tensor with a k-dimensional block (0 < k <= n), it stores how99 * Dn-1]. For i, j where 0 <= i < j < k, blockMap[i] < blockMap[j]. If not block sparse,108 * the dense tensor. First n elements represent the sparse tensor’s shape, and the last k111 * together specify the sparse indices along that dimension. The first pair of arguments
1542 * Computed bit vector is considered to be sparse.1571 * The offset value for sparse projections was added in %{NNAPILevel3}.
289 sparse: true,
1dictionary=main:it,locale=it,description=Italiano,date=1414726258, ...