Struct tract_core::ops::cnn::SumPool
source · Fields§
§pool_spec: PoolSpec§count_include_pad: bool§normalize: boolImplementations§
Trait Implementations§
source§impl Op for SumPool
impl Op for SumPool
fn name(&self) -> Cow<'_, str>
source§fn info(&self) -> TractResult<Vec<String>>
fn info(&self) -> TractResult<Vec<String>>
Short (one-line) strings giving hints on internal implementation or
important configuration details to be displayed in dumps. Read more
source§fn validation(&self) -> Validation
fn validation(&self) -> Validation
The kind of accuracy check that should be performed on operation when
testing them. Read more
source§fn op_families(&self) -> &'static [&'static str]
fn op_families(&self) -> &'static [&'static str]
Vector of short strings defining what families the op belongs too.
tract-core defines “core”, “mir”, “lir”. Read more
fn as_typed(&self) -> Option<&dyn TypedOp>
source§impl TypedOp for SumPool
impl TypedOp for SumPool
source§fn output_facts(&self, inputs: &[&TypedFact]) -> TractResult<TVec<TypedFact>>
fn output_facts(&self, inputs: &[&TypedFact]) -> TractResult<TVec<TypedFact>>
Deduce output facts from input facts.
fn invariants(
&self,
inputs: &[&TypedFact],
outputs: &[&TypedFact]
) -> TractResult<Invariants>
source§fn fuse(
&self,
_model: &TypedModel,
_node: &TypedNode
) -> TractResult<Option<TypedModelPatch>>
fn fuse(
&self,
_model: &TypedModel,
_node: &TypedNode
) -> TractResult<Option<TypedModelPatch>>
Fuse op after codegen to deal with local optimisations.
source§fn declutter_with_session(
&self,
session: &mut OptimizerSession<'_>,
model: &TypedModel,
node: &TypedNode
) -> TractResult<Option<TypedModelPatch>>
fn declutter_with_session(
&self,
session: &mut OptimizerSession<'_>,
model: &TypedModel,
node: &TypedNode
) -> TractResult<Option<TypedModelPatch>>
Declutter the op to the tract_core operator set as much as possible.
source§fn declutter(
&self,
model: &TypedModel,
node: &TypedNode
) -> TractResult<Option<TypedModelPatch>>
fn declutter(
&self,
model: &TypedModel,
node: &TypedNode
) -> TractResult<Option<TypedModelPatch>>
Declutter the op to the tract_core operator set as much as possible.
source§fn cost(&self, _inputs: &[&TypedFact]) -> TractResult<TVec<(Cost, TDim)>>
fn cost(&self, _inputs: &[&TypedFact]) -> TractResult<TVec<(Cost, TDim)>>
Computes a cost hint of the operation. Read more
fn suggested_axis_changes(&self) -> TractResult<TVec<(InOut, AxisOp)>>
fn change_axes(
&self,
model: &TypedModel,
node: &TypedNode,
io: InOut,
change: &AxisOp
) -> TractResult<Option<AxisChangeConsequence>>
fn slice_output(
&self,
model: &TypedModel,
node: &TypedNode,
patch: &mut TypedModelPatch,
slice_suffix: &str,
output_slot: usize,
axis: usize,
start: usize,
end: usize
) -> TractResult<Option<(OutletId, bool)>>
source§fn quantize(
&self,
model: &TypedModel,
node: &TypedNode,
dt: DatumType,
scale: f32,
zero_point: i32
) -> TractResult<Option<Box<dyn TypedOp>>>
fn quantize(
&self,
model: &TypedModel,
node: &TypedNode,
dt: DatumType,
scale: f32,
zero_point: i32
) -> TractResult<Option<Box<dyn TypedOp>>>
Transforms the op in an equivalent one, operating on dt (i8 or u8). Read more
source§fn concretize_dims(
&self,
source: &TypedModel,
node: &TypedNode,
target: &mut TypedModel,
mapping: &HashMap<OutletId, OutletId>,
values: &SymbolValues
) -> TractResult<TVec<OutletId>>
fn concretize_dims(
&self,
source: &TypedModel,
node: &TypedNode,
target: &mut TypedModel,
mapping: &HashMap<OutletId, OutletId>,
values: &SymbolValues
) -> TractResult<TVec<OutletId>>
Transform the op into by providing a value to one or more symbols.
source§fn codegen(
&self,
model: &TypedModel,
node: &TypedNode
) -> TractResult<Option<TypedModelPatch>>
fn codegen(
&self,
model: &TypedModel,
node: &TypedNode
) -> TractResult<Option<TypedModelPatch>>
Translate the op into the most efficient form possible for execution. Read more
Auto Trait Implementations§
impl RefUnwindSafe for SumPool
impl Send for SumPool
impl Sync for SumPool
impl Unpin for SumPool
impl UnwindSafe for SumPool
Blanket Implementations§
source§impl<T> Downcast for Twhere
T: Any,
impl<T> Downcast for Twhere
T: Any,
source§fn into_any(self: Box<T, Global>) -> Box<dyn Any + 'static, Global>
fn into_any(self: Box<T, Global>) -> Box<dyn Any + 'static, Global>
Convert
Box<dyn Trait> (where Trait: Downcast) to Box<dyn Any>. Box<dyn Any> can
then be further downcast into Box<ConcreteType> where ConcreteType implements Trait. Read moresource§fn into_any_rc(self: Rc<T>) -> Rc<dyn Any + 'static>
fn into_any_rc(self: Rc<T>) -> Rc<dyn Any + 'static>
Convert
Rc<Trait> (where Trait: Downcast) to Rc<Any>. Rc<Any> can then be
further downcast into Rc<ConcreteType> where ConcreteType implements Trait. Read moresource§fn as_any(&self) -> &(dyn Any + 'static)
fn as_any(&self) -> &(dyn Any + 'static)
Convert
&Trait (where Trait: Downcast) to &Any. This is needed since Rust cannot
generate &Any’s vtable from &Trait’s. Read moresource§fn as_any_mut(&mut self) -> &mut (dyn Any + 'static)
fn as_any_mut(&mut self) -> &mut (dyn Any + 'static)
Convert
&mut Trait (where Trait: Downcast) to &Any. This is needed since Rust cannot
generate &mut Any’s vtable from &mut Trait’s. Read more