For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).
Mojo function
set_batched_scale_factor
def set_batched_scale_factor[scales_dtype: DType, //, SF_VECTOR_SIZE: Int](scales_tensor: TileTensor[scales_dtype, Engine=scales_tensor.Engine, address_space=scales_tensor.address_space, linear_idx_type=scales_tensor.linear_idx_type], batch_idx: Int, row_idx: Int, col_idx: Int, scale_value: Scalar[scales_dtype])
Stores a scale factor into a 6D batched TileTensor at the given batch, row, and column.
TileTensor overload of set_batched_scale_factor that translates the
linear (batch_idx, row_idx, col_idx) coordinates into the 6D scale-factor
layout and stores scale_value there.
Parameters:
- scales_dtype (
DType): Element type of the scales tensor. - SF_VECTOR_SIZE (
Int): Number of elements each scale factor covers.
Args:
- scales_tensor (
TileTensor[scales_dtype, Engine=scales_tensor.Engine, address_space=scales_tensor.address_space, linear_idx_type=scales_tensor.linear_idx_type]): MutableTileTensor(flat rank == 6) holding the batched scale factors. - batch_idx (
Int): Batch index in the original (unscaled) tensor coordinates. - row_idx (
Int): Row index in the original (unscaled) tensor coordinates. - col_idx (
Int): Column index in the original (unscaled) tensor coordinates. - scale_value (
Scalar[scales_dtype]): Scale factor value to store.