Struct Mamba1 

pub struct Mamba1<B: Backend> {
    pub in_proj: Linear<B>,
    pub conv1d: Conv1d<B>,
    pub x_proj: Linear<B>,
    pub dt_proj: Linear<B>,
    pub a_log: Param<Tensor<B, 2>>,
    pub d: Param<Tensor<B, 1>>,
    pub out_proj: Linear<B>,
}

Fields

in_proj: Linear<B>

Input channel: d_model. Output channel: 2 * d_inner.

conv1d: Conv1d<B>

Input channel: d_inner. Output channel: d_inner.

x_proj: Linear<B>

Input channel: d_inner. Output channel: dt_rank + 2 * d_state.

dt_proj: Linear<B>

Input channel: dt_rank. Output channel: d_inner.

a_log: Param<Tensor<B, 2>>

Dims: [d_inner, d_state].

d: Param<Tensor<B, 1>>

Dims: [d_inner].

out_proj: Linear<B>

Input channel: d_inner. Output channel: d_model.

Implementations

impl<B: Backend> Mamba1<B>

pub fn step( &self, x: Tensor<B, 2>, cache: Mamba1Cache<B>, ) -> (Tensor<B, 2>, Mamba1Cache<B>)

Shapes

• Input [batch, d_model]
• Output [batch, d_model]

pub fn ss_step( &self, u: Tensor<B, 2>, cache: Mamba1Cache<B>, ) -> (Tensor<B, 2>, Mamba1Cache<B>)

Runs the SSM. See:

• Algorithm 2 in Section 3.2 from the Mamba paper;
• run_SSM(A, B, C, u) from The Annotated S4.

Shapes

• Input u [batch, d_inner]
• Output [batch, d_inner]

pub fn selective_scan_step( delta: Tensor<B, 2>, a: Tensor<B, 2>, b: Tensor<B, 2>, c: Tensor<B, 2>, d: Tensor<B, 1>, u: Tensor<B, 2>, cache: Mamba1Cache<B>, ) -> (Tensor<B, 2>, Mamba1Cache<B>)

Selective Scan.

Does selective scan algorithm. See:

• Section 2 State Space Models from the Mamba paper;
• Algorithm 2 in Section 3.2 from the Mamba paper;
• run_SSM(A, B, C, u) from The Annotated S4.

Shapes

• Input delta [batch, d_inner]
• Input a [d_inner, d_state]
• Input b [batch, d_state]
• Input c [batch, d_state]
• Input d [d_inner]
• Input u [batch, d_inner]
• Output [batch, d_inner]

impl<B: Backend> Mamba1<B>

pub fn forward(&self, x: Tensor<B, 3>) -> Tensor<B, 3>

See also Self::step.

Shapes

• Input [batch, sequence, d_model]
• Output [batch, sequence, d_model]

pub fn ss(&self, u: Tensor<B, 3>) -> Tensor<B, 3>

Shapes

• Input [batch, sequence, d_inner]
• Output [batch, sequence, d_inner]

pub fn selective_scan( delta: Tensor<B, 3>, a: Tensor<B, 2>, b: Tensor<B, 3>, c: Tensor<B, 3>, d: Tensor<B, 1>, u: Tensor<B, 3>, ) -> Tensor<B, 3>

Selective Scan.

Does selective scan algorithm. See:

• Section 2 State Space Models from the Mamba paper;
• Algorithm 2 in Section 3.2 from the Mamba paper;
• run_SSM(A, B, C, u) from The Annotated S4.

Shapes

• Input delta [sequence, batch, d_inner]
• Input a [d_inner, d_state]
• Input b [batch, sequence, d_state]
• Input c [sequence, batch, d_state]
• Input d [d_inner]
• Input u [batch, sequence, d_inner]
• Output [batch, sequence, d_inner]

Trait Implementations

impl<B> AutodiffModule<B> for Mamba1<B>

where B: AutodiffBackend + Backend, <B as AutodiffBackend>::InnerBackend: Backend,

type InnerModule = Mamba1<<B as AutodiffBackend>::InnerBackend>

Inner module without auto-differentiation.

fn valid(&self) -> Self::InnerModule

Returns the same module, but on the inner backend without auto-differentiation.

fn from_inner(module: Self::InnerModule) -> Self

Wraps an inner module back into an auto-diff module.

impl<B: Backend> Clone for Mamba1<B>

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<B: Debug + Backend> Debug for Mamba1<B>

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

Formats the value using the given formatter.

impl<B: Backend> Display for Mamba1<B>

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

Formats the value using the given formatter.

impl<B> HasAutodiffModule<B> for Mamba1<B::InnerBackend>

where B: AutodiffBackend + Backend, <B as AutodiffBackend>::InnerBackend: Backend,

type TrainModule = Mamba1<B>

The module with auto-differentiation.

impl<B: Backend> Module<B> for Mamba1<B>

type Record = Mamba1Record<B>

Type to save and load the module.

fn load_record(self, record: Self::Record) -> Self

Load the module state from a record.

fn into_record(self) -> Self::Record

Convert the module into a record containing the state.

fn num_params(&self) -> usize

Get the number of parameters the module has, including all of its sub-modules.

fn visit<Visitor: ModuleVisitor<B>>(&self, visitor: &mut Visitor)

Visit each tensor parameter in the module with a visitor.

fn map<Mapper: ModuleMapper<B>>(self, mapper: &mut Mapper) -> Self

Map each tensor parameter in the module with a mapper.

fn collect_devices(&self, devices: Devices<B>) -> Devices<B>

Return all the devices found in the underneath module tree added to the given vector without duplicates.

fn to_device(self, device: &B::Device) -> Self

Move the module and all of its sub-modules to the given device.

fn fork(self, device: &B::Device) -> Self

Fork the module and all of its sub-modules to the given device.

fn devices(&self) -> Vec<<B as BackendTypes>::Device>

Return all the devices found in the underneath module tree without duplicates.

fn no_grad(self) -> Self

Each tensor in the module tree will not require grad.

fn train<AB>(self) -> Self::TrainModule

where AB: AutodiffBackend<InnerBackend = B>, Self: HasAutodiffModule<AB>,

Move the module and all of its sub-modules to the autodiff backend.

fn quantize_weights(self, quantizer: &mut Quantizer) -> Self

Quantize the weights of the module.

impl<B: Backend> ModuleDisplay for Mamba1<B>

fn format(&self, passed_settings: DisplaySettings) -> String

Formats the module with provided display settings.

fn custom_settings(&self) -> Option<DisplaySettings>

Custom display settings for the module.

fn custom_content(&self, _content: Content) -> Option<Content>

Custom attributes for the module.

impl<B: Backend> ModuleDisplayDefault for Mamba1<B>

fn content(&self, content: Content) -> Option<Content>

Attributes of the module used for display purposes.

fn num_params(&self) -> usize

Gets the number of the parameters of the module.