pulsatrix/api/collate_8hpp_source.html

#pragma once


#include <functional>

#include <vector>


#include "pulsatrix/dataset.hpp"

#include "pulsatrix/device_backend.hpp"

#include "pulsatrix/tensor.hpp"


namespace pulsatrix {


struct Batch {

    std::vector<Tensor> fields;


    [[nodiscard]] int64_t size() const { return fields.empty() ? 0 : fields[0].shape().dim(0); }

};


using CollateFn = std::function<Batch(std::vector<Sample>, DeviceBackend*)>;


[[nodiscard]] Batch DefaultCollate(std::vector<Sample> samples, DeviceBackend* backend);


}  // namespace pulsatrix

pulsatrix::DeviceBackend
Vendor-agnostic compute/memory backend. CPUBackend, CUDABackend (Phase 1.5), and HIPBackend (Phase 1....
Definition device_backend.hpp:219

dataset.hpp
Random-access dataset abstraction – Sample, Dataset (size()/get()).

device_backend.hpp
Abstract interface isolating vendor-specific memory/compute operations from Tensor/ComputationGraph.

pulsatrix
Definition acquisition_functions.hpp:16

pulsatrix::DefaultCollate
Batch DefaultCollate(std::vector< Sample > samples, DeviceBackend *backend)
Stacks a list of samples into one Batch, field-by-field, via Tensor::Stack – pulsatrix's default Coll...

pulsatrix::CollateFn
std::function< Batch(std::vector< Sample >, DeviceBackend *)> CollateFn
A function assembling a list of Samples into one Batch – pulsatrix's analogue of PyTorch's collate_fn...
Definition collate.hpp:30

pulsatrix::Batch
One collated batch: one stacked Tensor per Sample field position.
Definition collate.hpp:17

pulsatrix::Batch::size
int64_t size() const
Number of samples in this batch – fields[0]'s leading dimension.
Definition collate.hpp:21

pulsatrix::Batch::fields
std::vector< Tensor > fields
Definition collate.hpp:18

tensor.hpp
N-dimensional tensor – owns a buffer via DeviceBackend*, RAII (Rule of Five).