Skip to main content

GatherQmm

Struct GatherQmm 

Source
#[non_exhaustive]
pub struct GatherQmm<'a> { pub transpose: bool, pub sorted_indices: bool, pub global_scale: Option<&'a Array>, }
Expand description

Options of gather_qmm, built with setters. GatherQmm::default() multiplies by the transposed weights (MLX’s default).

Fields (Non-exhaustive)§

This struct is marked as non-exhaustive
Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.
§transpose: bool

x · wᵀ (MLX’s default) rather than x · w.

§sorted_indices: bool

The right indices are sorted (rows grouped by expert), so one kernel can walk each expert’s weights once.

§global_scale: Option<&'a Array>

nvfp4 only: one float32 scale per expert (w.weight’s batch shape), multiplying the expert’s decoded weights by global_scale / (448 · 6). Needs the right indices.

Implementations§

Source§

impl<'a> GatherQmm<'a>

Source

pub fn groups(rows: usize, experts: usize) -> bool

Whether a gather of rows sorted index rows over experts weight batches takes the grouped kernel (each expert’s rows in tiles), rather than one product per row. Two gathers that take the same kernel give a row the same bits, whatever other rows are gathered with it: a caller batching gathers keeps each one’s bits when its rows and the batch’s agree here.

Source

pub fn transpose(self, transpose: bool) -> Self

Whether the weights are used transposed.

Source

pub fn sorted_indices(self, sorted_indices: bool) -> Self

Whether the right indices are sorted.

Source

pub fn global_scale(self, global_scale: &'a Array) -> Self

The per-expert nvfp4 global scale.

Trait Implementations§

Source§

impl<'a> Clone for GatherQmm<'a>

Source§

fn clone(&self) -> Self

Returns a duplicate of the value. Read more
1.0.0 (const: unstable) · Source§

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source. Read more
Source§

impl<'a> Copy for GatherQmm<'a>

Source§

impl<'a> Debug for GatherQmm<'a>

Source§

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter. Read more
Source§

impl Default for GatherQmm<'_>

Source§

fn default() -> Self

Returns the “default value” for a type. Read more

Auto Trait Implementations§

§

impl<'a> !RefUnwindSafe for GatherQmm<'a>

§

impl<'a> !UnwindSafe for GatherQmm<'a>

§

impl<'a> Freeze for GatherQmm<'a>

§

impl<'a> Send for GatherQmm<'a>

§

impl<'a> Sync for GatherQmm<'a>

§

impl<'a> Unpin for GatherQmm<'a>

§

impl<'a> UnsafeUnpin for GatherQmm<'a>

Blanket Implementations§

Source§

impl<T> Any for T
where T: 'static + ?Sized,

Source§

fn type_id(&self) -> TypeId

Gets the TypeId of self. Read more
Source§

impl<T> AutoreleaseSafe for T
where T: ?Sized,

Source§

impl<T> Borrow<T> for T
where T: ?Sized,

Source§

fn borrow(&self) -> &T

Immutably borrows from an owned value. Read more
Source§

impl<T> BorrowMut<T> for T
where T: ?Sized,

Source§

fn borrow_mut(&mut self) -> &mut T

Mutably borrows from an owned value. Read more
Source§

impl<ST, DT> CastableFrom<ST, Initialized, Initialized> for DT
where ST: ?Sized, DT: ?Sized,

Source§

impl<ST, DT> CastableFrom<ST, Uninit, Uninit> for DT
where ST: ?Sized, DT: ?Sized,

Source§

impl<T> CloneToUninit for T
where T: Clone,

Source§

unsafe fn clone_to_uninit(&self, dest: *mut u8)

🔬This is a nightly-only experimental API. (clone_to_uninit)
Performs copy-assignment from self to dest. Read more
Source§

impl<T> From<T> for T

Source§

fn from(t: T) -> T

Returns the argument unchanged.

Source§

impl<T, U> Into<U> for T
where U: From<T>,

Source§

fn into(self) -> U

Calls U::from(self).

That is, this conversion is whatever the implementation of From<T> for U chooses to do.

Source§

impl<T> IntoEither for T

Source§

fn into_either(self, into_left: bool) -> Either<Self, Self> ⓘ

Converts self into a Left variant of Either<Self, Self> if into_left is true. Converts self into a Right variant of Either<Self, Self> otherwise. Read more
Source§

fn into_either_with<F>(self, into_left: F) -> Either<Self, Self> ⓘ
where F: FnOnce(&Self) -> bool,

Converts self into a Left variant of Either<Self, Self> if into_left(&self) returns true. Converts self into a Right variant of Either<Self, Self> otherwise. Read more
Source§

impl<T> Pointable for T

Source§

const ALIGN: usize

The alignment of pointer.
Source§

type Init = T

The type for initializers.
Source§

unsafe fn init(init: <T as Pointable>::Init) -> usize

Initializes a with the given initializer. Read more
Source§

unsafe fn deref<'a>(ptr: usize) -> &'a T

Dereferences the given pointer. Read more
Source§

unsafe fn deref_mut<'a>(ptr: usize) -> &'a mut T

Mutably dereferences the given pointer. Read more
Source§

unsafe fn drop(ptr: usize)

Drops the object pointed to by the given pointer. Read more
Source§

impl<T> Read<Exclusive, BecauseExclusive> for T
where T: ?Sized,

Source§

impl<T> ToOwned for T
where T: Clone,

Source§

type Owned = T

The resulting type after obtaining ownership.
Source§

fn to_owned(&self) -> T

Creates owned data from borrowed data, usually by cloning. Read more
Source§

fn clone_into(&self, target: &mut T)

Uses borrowed data to replace owned data, usually by cloning. Read more
Source§

impl<T, U> TryFrom<U> for T
where U: Into<T>,

Source§

type Error = !

The type returned in the event of a conversion error.
Source§

fn try_from(value: U) -> Result<T, !>

Performs the conversion.
Source§

impl<T, U> TryInto<U> for T
where U: TryFrom<T>,

Source§

type Error = <U as TryFrom<T>>::Error

The type returned in the event of a conversion error.
Source§

fn try_into(self) -> Result<U, <U as TryFrom<T>>::Error>

Performs the conversion.