-
Notifications
You must be signed in to change notification settings - Fork 197
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
5 changed files
with
271 additions
and
35 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
238 changes: 238 additions & 0 deletions
238
cpp/include/raft/neighbors/detail/dataset_serialize.hpp
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,238 @@ | ||
| /* | ||
| * Copyright (c) 2024, NVIDIA CORPORATION. | ||
| * | ||
| * Licensed under the Apache License, Version 2.0 (the "License"); | ||
| * you may not use this file except in compliance with the License. | ||
| * You may obtain a copy of the License at | ||
| * | ||
| * http://www.apache.org/licenses/LICENSE-2.0 | ||
| * | ||
| * Unless required by applicable law or agreed to in writing, software | ||
| * distributed under the License is distributed on an "AS IS" BASIS, | ||
| * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. | ||
| * See the License for the specific language governing permissions and | ||
| * limitations under the License. | ||
| */ | ||
| #pragma once | ||
|
|
||
| #include "../dataset.hpp" | ||
|
|
||
| #include <raft/core/host_mdarray.hpp> | ||
| #include <raft/core/resources.hpp> | ||
| #include <raft/core/serialize.hpp> | ||
|
|
||
| #include <cuda_fp16.h> | ||
|
|
||
| #include <fstream> | ||
| #include <memory> | ||
|
|
||
| namespace raft::neighbors::detail { | ||
|
|
||
| using dataset_instance_tag = uint32_t; | ||
| constexpr dataset_instance_tag kSerializeEmptyDataset = 1; | ||
| constexpr dataset_instance_tag kSerializeStridedDataset = 2; | ||
| constexpr dataset_instance_tag kSerializeVPQDataset = 3; | ||
|
|
||
| template <typename IdxT> | ||
| void serialize(const raft::resources& res, std::ostream& os, const empty_dataset<IdxT>& dataset) | ||
| { | ||
| serialize_scalar(res, os, dataset.suggested_dim); | ||
| } | ||
|
|
||
| template <typename DataT, typename IdxT> | ||
| void serialize(const raft::resources& res, | ||
| std::ostream& os, | ||
| const strided_dataset<DataT, IdxT>& dataset) | ||
| { | ||
| serialize_scalar(res, os, dataset.n_rows()); | ||
| serialize_scalar(res, os, dataset.dim()); | ||
| serialize_scalar(res, os, dataset.stride()); | ||
| // Remove padding before saving the dataset | ||
| auto src = dataset.view(); | ||
| auto dst = make_host_mdarray<DataT, IdxT>(src.extents()); | ||
| RAFT_CUDA_TRY(cudaMemcpy2DAsync(dst.data_handle(), | ||
| sizeof(DataT) * dst.extent(1), | ||
| src.data_handle(), | ||
| sizeof(DataT) * src.stride(0), | ||
| sizeof(DataT) * dst.extent(1), | ||
| src.extent(0), | ||
| cudaMemcpyDefault, | ||
| resource::get_cuda_stream(res))); | ||
| resource::sync_stream(res); | ||
| serialize_mdspan(res, os, dst.view()); | ||
| } | ||
|
|
||
| template <typename MathT, typename IdxT> | ||
| void serialize(const raft::resources& res, | ||
| std::ostream& os, | ||
| const vpq_dataset<MathT, IdxT>& dataset) | ||
| { | ||
| serialize_scalar(res, os, dataset.n_rows()); | ||
| serialize_scalar(res, os, dataset.dim()); | ||
| serialize_scalar(res, os, dataset.vq_n_centers()); | ||
| serialize_scalar(res, os, dataset.pq_n_centers()); | ||
| serialize_scalar(res, os, dataset.pq_len()); | ||
| serialize_scalar(res, os, dataset.encoded_row_length()); | ||
| serialize_mdspan(res, os, make_const_mdspan(dataset.vq_code_book.view())); | ||
| serialize_mdspan(res, os, make_const_mdspan(dataset.pq_code_book.view())); | ||
| serialize_mdspan(res, os, make_const_mdspan(dataset.data.view())); | ||
| } | ||
|
|
||
| template <typename IdxT> | ||
| void serialize(const raft::resources& res, std::ostream& os, const dataset<IdxT>& dataset) | ||
| { | ||
| if (auto x = dynamic_cast<const empty_dataset<IdxT>*>(&dataset); x != nullptr) { | ||
| serialize_scalar(res, os, kSerializeEmptyDataset); | ||
| return serialize(res, os, *x); | ||
| } | ||
| if (auto x = dynamic_cast<const strided_dataset<float, IdxT>*>(&dataset); x != nullptr) { | ||
| serialize_scalar(res, os, kSerializeStridedDataset); | ||
| serialize_scalar(res, os, CUDA_R_32F); | ||
| return serialize(res, os, *x); | ||
| } | ||
| if (auto x = dynamic_cast<const strided_dataset<half, IdxT>*>(&dataset); x != nullptr) { | ||
| serialize_scalar(res, os, kSerializeStridedDataset); | ||
| serialize_scalar(res, os, CUDA_R_16F); | ||
| return serialize(res, os, *x); | ||
| } | ||
| if (auto x = dynamic_cast<const strided_dataset<int8_t, IdxT>*>(&dataset); x != nullptr) { | ||
| serialize_scalar(res, os, kSerializeStridedDataset); | ||
| serialize_scalar(res, os, CUDA_R_8I); | ||
| return serialize(res, os, *x); | ||
| } | ||
| if (auto x = dynamic_cast<const strided_dataset<uint8_t, IdxT>*>(&dataset); x != nullptr) { | ||
| serialize_scalar(res, os, kSerializeStridedDataset); | ||
| serialize_scalar(res, os, CUDA_R_8U); | ||
| return serialize(res, os, *x); | ||
| } | ||
| if (auto x = dynamic_cast<const vpq_dataset<float, IdxT>*>(&dataset); x != nullptr) { | ||
| serialize_scalar(res, os, kSerializeVPQDataset); | ||
| serialize_scalar(res, os, CUDA_R_32F); | ||
| return serialize(res, os, *x); | ||
| } | ||
| if (auto x = dynamic_cast<const vpq_dataset<half, IdxT>*>(&dataset); x != nullptr) { | ||
| serialize_scalar(res, os, kSerializeVPQDataset); | ||
| serialize_scalar(res, os, CUDA_R_16F); | ||
| return serialize(res, os, *x); | ||
| } | ||
| RAFT_FAIL("unsupported dataset type."); | ||
| } | ||
|
|
||
| template <typename IdxT> | ||
| void deserialize(raft::resources const& res, | ||
| std::istream& is, | ||
| std::unique_ptr<empty_dataset<IdxT>>& out) | ||
| { | ||
| auto suggested_dim = deserialize_scalar<uint32_t>(res, is); | ||
| return std::make_unique<empty_dataset<IdxT>>(suggested_dim).swap(out); | ||
| } | ||
|
|
||
| template <typename DataT, typename IdxT> | ||
| void deserialize(raft::resources const& res, | ||
| std::istream& is, | ||
| std::unique_ptr<strided_dataset<DataT, IdxT>>& out) | ||
| { | ||
| using out_mdarray_type = device_mdarray<DataT, matrix_extent<IdxT>, layout_stride>; | ||
| using out_layout_type = typename out_mdarray_type::layout_type; | ||
| using out_container_policy_type = typename out_mdarray_type::container_policy_type; | ||
| using out_owning_type = owning_dataset<DataT, IdxT, out_layout_type, out_container_policy_type>; | ||
|
|
||
| auto n_rows = deserialize_scalar<IdxT>(res, is); | ||
| auto dim = deserialize_scalar<uint32_t>(res, is); | ||
| auto stride = deserialize_scalar<uint32_t>(res, is); | ||
| auto out_extents = make_extents<IdxT>(n_rows, dim); | ||
| auto out_layout = make_strided_layout(out_extents, std::array<IdxT, 2>{stride, 1}); | ||
| auto out_array = out_mdarray_type{res, out_layout, out_container_policy_type{}}; | ||
| auto host_arrray = make_host_mdarray<DataT, IdxT>(out_extents); | ||
| deserialize_mdspan(res, is, host_arrray.view()); | ||
| RAFT_CUDA_TRY(cudaMemcpy2DAsync(out_array.data_handle(), | ||
| sizeof(DataT) * stride, | ||
| host_arrray.data_handle(), | ||
| sizeof(DataT) * dim, | ||
| sizeof(DataT) * dim, | ||
| n_rows, | ||
| cudaMemcpyDefault, | ||
| resource::get_cuda_stream(res))); | ||
| return std::unique_ptr<strided_dataset<DataT, IdxT>>{ | ||
| new out_owning_type{std::move(out_array), out_layout}} | ||
| .swap(out); | ||
| } | ||
|
|
||
| template <typename MathT, typename IdxT> | ||
| void deserialize(raft::resources const& res, | ||
| std::istream& is, | ||
| std::unique_ptr<vpq_dataset<MathT, IdxT>>& out) | ||
| { | ||
| auto n_rows = deserialize_scalar<IdxT>(res, is); | ||
| auto dim = deserialize_scalar<uint32_t>(res, is); | ||
| auto vq_n_centers = deserialize_scalar<uint32_t>(res, is); | ||
| auto pq_n_centers = deserialize_scalar<uint32_t>(res, is); | ||
| auto pq_len = deserialize_scalar<uint32_t>(res, is); | ||
| auto encoded_row_length = deserialize_scalar<uint32_t>(res, is); | ||
|
|
||
| auto vq_code_book = make_device_matrix<MathT, uint32_t, row_major>(res, vq_n_centers, dim); | ||
| auto pq_code_book = make_device_matrix<MathT, uint32_t, row_major>(res, pq_n_centers, pq_len); | ||
| auto data = make_device_matrix<uint8_t, IdxT, row_major>(res, n_rows, encoded_row_length); | ||
|
|
||
| deserialize_mdspan(res, is, vq_code_book.view()); | ||
| deserialize_mdspan(res, is, pq_code_book.view()); | ||
| deserialize_mdspan(res, is, data.view()); | ||
|
|
||
| return std::unique_ptr<vpq_dataset<MathT, IdxT>>{ | ||
| new vpq_dataset{std::move(vq_code_book), std::move(pq_code_book), std::move(data)}} | ||
| .swap(out); | ||
| } | ||
|
|
||
| template <typename IdxT> | ||
| void deserialize(raft::resources const& res, std::istream& is, std::unique_ptr<dataset<IdxT>>& out) | ||
| { | ||
| switch (deserialize_scalar<dataset_instance_tag>(res, is)) { | ||
| case kSerializeEmptyDataset: { | ||
| std::unique_ptr<empty_dataset<IdxT>> p; | ||
| deserialize(res, is, p); | ||
| return upcast_dataset_ptr(std::move(p)).swap(out); | ||
| } | ||
| case kSerializeStridedDataset: | ||
| switch (deserialize_scalar<cudaDataType_t>(res, is)) { | ||
| case CUDA_R_32F: { | ||
| std::unique_ptr<strided_dataset<float, IdxT>> p; | ||
| deserialize(res, is, p); | ||
| return upcast_dataset_ptr(std::move(p)).swap(out); | ||
| } | ||
| case CUDA_R_16F: { | ||
| std::unique_ptr<strided_dataset<half, IdxT>> p; | ||
| deserialize(res, is, p); | ||
| return upcast_dataset_ptr(std::move(p)).swap(out); | ||
| } | ||
| case CUDA_R_8I: { | ||
| std::unique_ptr<strided_dataset<int8_t, IdxT>> p; | ||
| deserialize(res, is, p); | ||
| return upcast_dataset_ptr(std::move(p)).swap(out); | ||
| } | ||
| case CUDA_R_8U: { | ||
| std::unique_ptr<strided_dataset<uint8_t, IdxT>> p; | ||
| deserialize(res, is, p); | ||
| return upcast_dataset_ptr(std::move(p)).swap(out); | ||
| } | ||
| default: break; | ||
| } | ||
| case kSerializeVPQDataset: | ||
| switch (deserialize_scalar<cudaDataType_t>(res, is)) { | ||
| case CUDA_R_32F: { | ||
| std::unique_ptr<vpq_dataset<float, IdxT>> p; | ||
| deserialize(res, is, p); | ||
| return upcast_dataset_ptr(std::move(p)).swap(out); | ||
| } | ||
| case CUDA_R_16F: { | ||
| std::unique_ptr<vpq_dataset<half, IdxT>> p; | ||
| deserialize(res, is, p); | ||
| return upcast_dataset_ptr(std::move(p)).swap(out); | ||
| } | ||
| default: break; | ||
| } | ||
| default: break; | ||
| } | ||
| RAFT_FAIL("Failed to deserialize dataset: unsupported combination of instance tags."); | ||
| } | ||
|
|
||
| } // namespace raft::neighbors::detail |