pulsatrix 1.1.0
Loading...
Searching...
No Matches
pulsatrix::AdaptiveAvgPool2DModule Class Reference

Pools a rank-4 (N, C, H, W) input to (N, C, out_h, out_w), as AdaptiveAvgPool2d((out_h, out_w)): ResNet's global pooling is (1, 1), VGG's is (7, 7). More...

#include <adaptive_avg_pool2d_module.hpp>

Inheritance diagram for pulsatrix::AdaptiveAvgPool2DModule:
Collaboration diagram for pulsatrix::AdaptiveAvgPool2DModule:

Public Member Functions

 AdaptiveAvgPool2DModule (int64_t out_h, int64_t out_w, DeviceBackend *backend, float eps=1e-6f)
 
Tensor backward (const Tensor &grad_output) override
 Computes the gradient w.r.t. this module's input, given the gradient w.r.t. its output. Must be called after forward() – uses state cached from that call.
 
Tensor propagate_relevance (const Tensor &relevance_out, const LRPRuleConfig &config) override
 AvgPool2DModule's z-rule over each window.
 
OpType op_type () const override
 This module's operation-category tag, for ComputationGraph node tagging.
 
std::optional< DeviceType > compute_device () const override
 The device this module computes on, so forward() can reject an input on another device before any kernel sees it (roadmap FND-8).
 
int64_t out_h () const
 
int64_t out_w () const
 
- Public Member Functions inherited from pulsatrix::Module
virtual ~Module ()=default
 
Tensor forward (const Tensor &input)
 Runs this module's forward computation.
 
std::pair< Tensor, NodeId > forward_traced (const Tensor &input, NodeId input_node, ComputationGraph &graph, Autograd &autograd)
 Runs forward() while also registering a ComputationGraph node (tagged with this module's op_type(), parented to input_node) and wiring an Autograd backward function that reuses this module's own backward() – the opt-in traced/explainable path, per Phase 2 Mission 0.
 
virtual bool supports_lrp_rule (LRPRule rule) const
 Whether propagate_relevance() implements rule (no silent fallback: callers such as ExplainerContext::relevance_pass() throw rather than run a module on a rule it does not implement).
 
virtual std::vector< NamedParamRef > named_parameters ()
 This module's trainable parameters, each with its hierarchical name – the one place a module declares its parameters (roadmap FND-1).
 
virtual std::vector< NamedBufferRef > named_buffers ()
 This module's buffers: state a checkpoint must save that no optimizer updates.
 
virtual std::vector< NamedModuleRef > named_children ()
 This module's direct sub-modules, each under the prefix its parameters carry in named_parameters(), in the same order (TRN-7: how ApplyLora() finds a model's Linear layers). Default: none. NamedModules() walks the whole tree.
 
virtual std::vector< ParamRef > parameters ()
 This module's trainable parameters and their gradients, for an optimizer to update uniformly across module types.
 
void set_requires_grad (bool requires_grad, const std::string &prefix="")
 Freezes (false) or unfreezes (true) parameters by name (roadmap FND-2).
 
virtual void set_training (bool training)
 Sets this module's training/eval mode. Defaults to training (matches every mainstream framework's Module default).
 
bool is_training () const
 Whether this module is currently in training mode.
 
virtual void release_activations ()
 Frees the tensors forward() kept for backward() and the relevance passes; containers release their children too. Inference that only needs the output calls it after each forward, so a deep model holds one layer's activations at a time (a 1024-token ESM-2 650M pass otherwise keeps over 10 GB of them).
 

Protected Member Functions

Tensor forward_impl (const Tensor &input) override
 

Detailed Description

Pools a rank-4 (N, C, H, W) input to (N, C, out_h, out_w), as AdaptiveAvgPool2d((out_h, out_w)): ResNet's global pooling is (1, 1), VGG's is (7, 7).

Note
Only sizes that divide evenly are supported (H divisible by out_h, W by out_w): the windows are then the non-overlapping (H / out_h, W / out_w) of AvgPool2DModule, with its backward and its LRP z-rule. PyTorch's uneven windows (overlapping, of different sizes) are refused rather than approximated.

Constructor & Destructor Documentation

◆ AdaptiveAvgPool2DModule()

pulsatrix::AdaptiveAvgPool2DModule::AdaptiveAvgPool2DModule ( int64_t  out_h,
int64_t  out_w,
DeviceBackend *  backend,
float  eps = 1e-6f 
)
Exceptions
std::invalid_argumentif out_h or out_w < 1.

Member Function Documentation

◆ backward()

Tensor pulsatrix::AdaptiveAvgPool2DModule::backward ( const Tensor &  grad_output)
overridevirtual

Computes the gradient w.r.t. this module's input, given the gradient w.r.t. its output. Must be called after forward() – uses state cached from that call.

Parameters
grad_outputGradient w.r.t. this module's output.
Returns
Gradient w.r.t. this module's input.
Note
Promoted to the base class in Phase 2 Mission 0 – every existing subclass (LinearModule/ReluModule/Conv2DModule) already implemented this exact signature independently; making it virtual lets graph-wiring code (Module::forward_traced) call it polymorphically through a Module* without knowing the concrete subclass, the same way propagate_relevance already works.

Implements pulsatrix::Module.

◆ compute_device()

std::optional< DeviceType > pulsatrix::AdaptiveAvgPool2DModule::compute_device ( ) const
inlineoverridevirtual

The device this module computes on, so forward() can reject an input on another device before any kernel sees it (roadmap FND-8).

Returns
std::nullopt (the default) skips the check: a container whose layers check their own inputs, or a user module written before this existed. Every in-tree layer returns its device; EmbeddingModule doesn't, because it reads its indices through their own backend and so accepts them from any device.

Reimplemented from pulsatrix::Module.

◆ forward_impl()

Tensor pulsatrix::AdaptiveAvgPool2DModule::forward_impl ( const Tensor &  input)
overrideprotectedvirtual
Exceptions
std::invalid_argumentunless the input is rank 4 with H and W divisible by the output size.

Implements pulsatrix::Module.

◆ op_type()

OpType pulsatrix::AdaptiveAvgPool2DModule::op_type ( ) const
inlineoverridevirtual

This module's operation-category tag, for ComputationGraph node tagging.

Returns
This module's OpType (see op_type.hpp's closed set).
Note
Added in Phase 2 Mission 0 alongside backward() – lets graph-wiring code tag nodes generically through a Module* rather than switching on concrete subclass.

Implements pulsatrix::Module.

◆ out_h()

int64_t pulsatrix::AdaptiveAvgPool2DModule::out_h ( ) const
inline

◆ out_w()

int64_t pulsatrix::AdaptiveAvgPool2DModule::out_w ( ) const
inline

◆ propagate_relevance()

Tensor pulsatrix::AdaptiveAvgPool2DModule::propagate_relevance ( const Tensor &  relevance_out,
const LRPRuleConfig &  config 
)
overridevirtual

AvgPool2DModule's z-rule over each window.

Implements pulsatrix::Module.


The documentation for this class was generated from the following file: