BatchedMatmul.left_gradient: def of bendtorch-core@0.1.0.0
Import: import bendtorch-core@0.1.0.0/src/tensor/batched_matmul.bend as BatchedMatmul
Lines 384-389 of bendtorch-core@0.1.0.0/src/tensor/batched_matmul.bend: https://jend.leeeet.dev/read.txt?path=bendtorch-core@0.1.0.0/src/tensor/batched_matmul.bend&start_line=384&end_line=389

import Base
import bendtorch-core@0.1.0.0/src/shape/core.bend as Shape
import bendtorch-core@0.1.0.0/src/tensor/core.bend as Core

# The left operand's gradient over the output's batch, g @ right^T, of
# dimensions `left_dims(plan)`: g has the output's shape and right the right
# operand's. A broadcast left operand still has to be summed to its shape.
def left_gradient(+plan: Plan, g: Core.Dynamic, right: Core.Dynamic) ->
  Result<&2, &1, Core.Error, Core.Dynamic & Core.Dynamic & Core.Dynamic>:
  gradient_shaped(left_layout(plan), g, right, left_dims(plan), Shape.construct(left_dims(plan)))

Uses (9):
  https://jend.leeeet.dev/definition.txt?path=Base/base.bend&name=Pair
  https://jend.leeeet.dev/definition.txt?path=Base/base.bend&name=Result
  https://jend.leeeet.dev/definition.txt?path=bendtorch-core@0.1.0.0/src/shape/core.bend&name=construct
  https://jend.leeeet.dev/definition.txt?path=bendtorch-core@0.1.0.0/src/tensor/batched_matmul.bend&name=Plan
  https://jend.leeeet.dev/definition.txt?path=bendtorch-core@0.1.0.0/src/tensor/batched_matmul.bend&name=gradient_shaped
  https://jend.leeeet.dev/definition.txt?path=bendtorch-core@0.1.0.0/src/tensor/batched_matmul.bend&name=left_dims
  https://jend.leeeet.dev/definition.txt?path=bendtorch-core@0.1.0.0/src/tensor/batched_matmul.bend&name=left_layout
  https://jend.leeeet.dev/definition.txt?path=bendtorch-core@0.1.0.0/src/tensor/core.bend&name=Dynamic
  https://jend.leeeet.dev/definition.txt?path=bendtorch-core@0.1.0.0/src/tensor/core.bend&name=Error