flashinfer.cake_causal_conv1d_det_train_kcp.CakeCausalConv1dDetKcpMultiFunction

class flashinfer.cake_causal_conv1d_det_train_kcp.CakeCausalConv1dDetKcpMultiFunction(*args, **kwargs)

Autograd wrapper of several convolutions with one halo exchange per direction: apply(ctx, graph, handle, n, *xs, *weights[, cooperative]) returns the n outputs; the forward saves (*xs, *weights, *hists); the backward runs cake_causal_conv1d_det_train_kcp_multi_backward() (the optional trailing cooperative: its launch form, None or a bool). handle: one recompute handle for the call, whose "conv" entry holds the n pools (the saved set’s own tensors): the recompute binds them and exchanges nothing. Refused with graph replay (keyword or knob) and inside a CUDA-graph capture, like the single call.

__init__(*args, **kwargs)

Methods

__init__(*args, **kwargs)

apply

backward(fn_ctx, *grad_ys)

Define a formula for differentiating the operation with backward mode automatic differentiation.

forward(fn_ctx, ctx, graph, handle, n, *tensors)

Define the forward of the custom autograd Function.

jvp(ctx, *grad_inputs)

Define a formula for differentiating the operation with forward mode automatic differentiation.

mark_dirty(*args)

Mark given tensors as modified in an in-place operation.

mark_non_differentiable(*args)

Mark outputs as non-differentiable.

mark_shared_storage(*pairs)

maybe_clear_saved_tensors

name

register_hook

register_prehook

save_for_backward(*tensors)

Save given tensors for a future call to backward().

save_for_forward(*tensors)

Save given tensors for a future call to jvp().

set_materialize_grads(value)

Set whether to materialize grad tensors.

set_output_grad_dtype(*dtypes)

Declare the gradient dtype for each of this Function's outputs.

setup_context(ctx, inputs, output)

There are two ways to define the forward pass of an autograd.Function.

vjp(ctx, *grad_outputs)

Define a formula for differentiating the operation with backward mode automatic differentiation.

vmap(info, in_dims, *args)

Define the behavior for this autograd.Function underneath torch.vmap().

Attributes

boxed_grads_call

clear_saved_tensors_on_access

Bool that specifies if backward should receive grads as a single mutable list argument instead of individual args in an immutable tuple.

dirty_tensors

generate_vmap_rule

materialize_grads

metadata

needs_input_grad

next_functions

non_differentiable

output_grad_dtypes

requires_grad

saved_for_forward

saved_tensors

saved_variables

to_save