Merge branch 'branch-25.02' into rhdong/bitset-to-csr

CHANGELOG.md

@@ -1,3 +1,42 @@
+# raft 24.12.00 (11 Dec 2024)
+
+## 🚨 Breaking Changes
+
+- Do not initialize the pinned mdarray at construction time ([#2478](https://github.com/rapidsai/raft/pull/2478)) [@achirkin](https://github.com/achirkin)
+
+## 🐛 Bug Fixes
+
+- Skip gtests for new lanczos solver when CUDA version is 11.4 or below. ([#2520](https://github.com/rapidsai/raft/pull/2520)) [@cjnolet](https://github.com/cjnolet)
+- Switch `assert` to `static_assert` ([#2510](https://github.com/rapidsai/raft/pull/2510)) [@divyegala](https://github.com/divyegala)
+- Revert use of new Lanczos solver in spectral clustering ([#2507](https://github.com/rapidsai/raft/pull/2507)) [@lowener](https://github.com/lowener)
+- Put a ceiling on cuda-python ([#2486](https://github.com/rapidsai/raft/pull/2486)) [@bdice](https://github.com/bdice)
+- Don't presume pointers location infers usability. ([#2480](https://github.com/rapidsai/raft/pull/2480)) [@robertmaynard](https://github.com/robertmaynard)
+- Use Python for sccache hit rate computation. ([#2474](https://github.com/rapidsai/raft/pull/2474)) [@bdice](https://github.com/bdice)
+- Allow compilation with CUDA 12.6.1 ([#2469](https://github.com/rapidsai/raft/pull/2469)) [@robertmaynard](https://github.com/robertmaynard)
+
+## 🚀 New Features
+
+- [FEA] Lanczos solver v2 ([#2481](https://github.com/rapidsai/raft/pull/2481)) [@lowener](https://github.com/lowener)
+
+## 🛠️ Improvements
+
+- Skip gtests for Rmat Lanczos tests with cuda <= 11.4 ([#2525](https://github.com/rapidsai/raft/pull/2525)) [@benfred](https://github.com/benfred)
+- Upgrade to latest cutlass version ([#2503](https://github.com/rapidsai/raft/pull/2503)) [@vyasr](https://github.com/vyasr)
+- Removing some left over places where implicit instantiations were being ignored in headers ([#2501](https://github.com/rapidsai/raft/pull/2501)) [@cjnolet](https://github.com/cjnolet)
+- Remove leftover template project code. ([#2500](https://github.com/rapidsai/raft/pull/2500)) [@bdice](https://github.com/bdice)
+- 2412 remove libraft vss instantiations ([#2498](https://github.com/rapidsai/raft/pull/2498)) [@cjnolet](https://github.com/cjnolet)
+- Remove raft-ann-bench ([#2497](https://github.com/rapidsai/raft/pull/2497)) [@cjnolet](https://github.com/cjnolet)
+- Pin FAISS Version for raft-ann-bench ([#2496](https://github.com/rapidsai/raft/pull/2496)) [@tarang-jain](https://github.com/tarang-jain)
+- enforce wheel size limits and README formatting in CI, put a ceiling on Cython dependency ([#2490](https://github.com/rapidsai/raft/pull/2490)) [@jameslamb](https://github.com/jameslamb)
+- Do not initialize the pinned mdarray at construction time ([#2478](https://github.com/rapidsai/raft/pull/2478)) [@achirkin](https://github.com/achirkin)
+- Use environment variables in cache hit rate computation. ([#2475](https://github.com/rapidsai/raft/pull/2475)) [@bdice](https://github.com/bdice)
+- devcontainer: replace `VAULT_HOST` with `AWS_ROLE_ARN` ([#2472](https://github.com/rapidsai/raft/pull/2472)) [@jjacobelli](https://github.com/jjacobelli)
+- print sccache stats in builds ([#2470](https://github.com/rapidsai/raft/pull/2470)) [@jameslamb](https://github.com/jameslamb)
+- make package installations in CI stricter ([#2467](https://github.com/rapidsai/raft/pull/2467)) [@jameslamb](https://github.com/jameslamb)
+- Prune workflows based on changed files ([#2466](https://github.com/rapidsai/raft/pull/2466)) [@KyleFromNVIDIA](https://github.com/KyleFromNVIDIA)
+- Merge branch-24.10 into branch-24.12 ([#2461](https://github.com/rapidsai/raft/pull/2461)) [@jameslamb](https://github.com/jameslamb)
+- Update all rmm imports to use pylibrmm/librmm ([#2451](https://github.com/rapidsai/raft/pull/2451)) [@Matt711](https://github.com/Matt711)
+
 # raft 24.10.00 (9 Oct 2024)
 
 ## 🚨 Breaking Changes

cpp/bench/prims/sparse/bitmap_to_csr.cu

@@ -71,7 +71,7 @@ struct BitmapToCsrBench : public fixture {
   index_t create_sparse_matrix(index_t m, index_t n, float sparsity, std::vector<bitmap_t>& bitmap)
   {
     index_t total    = static_cast<index_t>(m * n);
-    index_t num_ones = static_cast<index_t>((total * 1.0f) * sparsity);
+    index_t num_ones = static_cast<index_t>((total * 1.0f) * (1.0f - sparsity));
     index_t res      = num_ones;
 
     for (auto& item : bitmap) {
@@ -141,7 +141,27 @@ const std::vector<bench_param<index_t>> getInputs()
   };
 
   const std::vector<TestParams> params_group = raft::util::itertools::product<TestParams>(
-    {index_t(10), index_t(1024)}, {index_t(1024 * 1024)}, {0.01f, 0.1f, 0.2f, 0.5f});
+    {index_t(10), index_t(1024)}, {index_t(1024 * 1024)}, {0.99f, 0.9f, 0.8f, 0.5f});
+
+  param_vec.reserve(params_group.size());
+  for (TestParams params : params_group) {
+    param_vec.push_back(bench_param<index_t>({params.m, params.n, params.sparsity}));
+  }
+  return param_vec;
+}
+
+template <typename index_t = int64_t>
+const std::vector<bench_param<index_t>> getLargeInputs()
+{
+  std::vector<bench_param<index_t>> param_vec;
+  struct TestParams {
+    index_t m;
+    index_t n;
+    float sparsity;
+  };
+
+  const std::vector<TestParams> params_group = raft::util::itertools::product<TestParams>(
+    {index_t(1), index_t(100)}, {index_t(100 * 1000000)}, {0.95f, 0.99f});
 
   param_vec.reserve(params_group.size());
   for (TestParams params : params_group) {
@@ -153,4 +173,6 @@ const std::vector<bench_param<index_t>> getInputs()
 RAFT_BENCH_REGISTER((BitmapToCsrBench<uint32_t, int, float>), "", getInputs<int>());
 RAFT_BENCH_REGISTER((BitmapToCsrBench<uint64_t, int, double>), "", getInputs<int>());
 
+RAFT_BENCH_REGISTER((BitmapToCsrBench<uint32_t, int64_t, float>), "", getLargeInputs<int64_t>());
+
 }  // namespace raft::bench::sparse

cpp/include/raft/random/detail/rmat_rectangular_generator.cuh

@@ -151,15 +151,16 @@ RAFT_KERNEL rmat_gen_kernel(IdxT* out,
   raft::random::PCGenerator gen{r.seed, r.base_subsequence + idx, 0};
   auto min_scale = min(r_scale, c_scale);
   IdxT i         = 0;
-  for (; i < min_scale; ++i) {
-    gen_and_update_bits(src_id, dst_id, a, a + b, a + b + c, r_scale, c_scale, i, gen);
-  }
-  for (; i < r_scale; ++i) {
-    gen_and_update_bits(src_id, dst_id, a + b, a + b, ProbT(1), r_scale, c_scale, i, gen);
-  }
-  for (; i < c_scale; ++i) {
-    gen_and_update_bits(src_id, dst_id, a + c, ProbT(1), ProbT(1), r_scale, c_scale, i, gen);
+  // Whether we have more rows than columns.
+  const bool more_rows = r_scale > c_scale;
+
+  for (; i < max_scale; ++i) {
+    ProbT A   = (i < min_scale) ? a : (more_rows ? a + b : a + c);
+    ProbT AB  = (i < min_scale) ? a + b : (more_rows ? a + b : ProbT(1));
+    ProbT ABC = (i < min_scale) ? a + b + c : ProbT(1);
+    gen_and_update_bits(src_id, dst_id, A, AB, ABC, r_scale, c_scale, i, gen);
   }
+
   store_ids(out, out_src, out_dst, src_id, dst_id, idx, n_edges);
 }