Skip to content

GitLab

Commit 14048a3c authored by justheuristic's avatar justheuristic
Browse files

safer cast

parent 5b169f18
Show whitespace changes
Inline Side-by-side
Showing with 2 additions and 1 deletion
bitsandbytes/autograd/_functions.py
View file @ 14048a3c
...@@ -371,7 +371,7 @@ class MatMul8bitLt(torch.autograd.Function): ...@@ -371,7 +371,7 @@ class MatMul8bitLt(torch.autograd.Function):
gradB32, SgradB32 = F.igemmlt(C32grad, CxAt, Sgrad, SAt) gradB32, SgradB32 = F.igemmlt(C32grad, CxAt, Sgrad, SAt)
grad_B = F.mm_dequant(gradB32, SgradB32, SCgradt, SCAt).to(ctx.dtype_B) grad_B = F.mm_dequant(gradB32, SgradB32, SCgradt, SCAt).to(ctx.dtype_B)
if state.threshold > 0.0 and subA is not None: if state.threshold > 0.0 and subA is not None:
grad_B[:, idx].addmm_(grad_output.t(), subA) grad_B[:, idx] += torch.mm(grad_output.t(), subA)
if req_gradA: if req_gradA:
if state.CBt is not None: if state.CBt is not None:
...@@ -384,6 +384,7 @@ class MatMul8bitLt(torch.autograd.Function): ...@@ -384,6 +384,7 @@ class MatMul8bitLt(torch.autograd.Function):
grad_A = F.mm_dequant(gradA32, SgradA32, SCgrad, state.SCBt).view(ctx.grad_shape).to(ctx.dtype_A) grad_A = F.mm_dequant(gradA32, SgradA32, SCgrad, state.SCBt).view(ctx.grad_shape).to(ctx.dtype_A)
elif state.CB is not None: elif state.CB is not None:
raise NotImplementedError("WIP")
CB = state.CB.to(ctx.dtype_B) CB = state.CB.to(ctx.dtype_B)
CB.mul_(state.SCB.unsqueeze(1).div_(127.0).to(CB.dtype)) CB.mul_(state.SCB.unsqueeze(1).div_(127.0).to(CB.dtype))
grad_A = torch.matmul(grad_output, CB).view(ctx.grad_shape).to(ctx.dtype_A) grad_A = torch.matmul(grad_output, CB).view(ctx.grad_shape).to(ctx.dtype_A)
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment