feat: Implement expt_0726

Signed-off-by: KunoiSayami <[email protected]>

2 files changed, 110 insertions, 0 deletions

diff --git a/CMakeLists.txt b/CMakeLists.txt
index bec9954..858cea0 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -81,3 +81,6 @@ set_target_properties(work PROPERTIES CUDA_ARCHITECTURES "75")
 add_executable(expt_0722 expt_0722.cpp)
 set_target_properties(expt_0722 PROPERTIES LINKER_LANGUAGE CXX)
 
+add_executable(expt_0726 expt_0726.cpp)
+set_target_properties(expt_0726 PROPERTIES LINKER_LANGUAGE CXX)
+
diff --git a/expt_0726.cpp b/expt_0726.cpp
new file mode 100644
index 0000000..5b37df3
--- /dev/null
+++ b/expt_0726.cpp
@@ -0,0 +1,107 @@
+#include <algorithm>
+#include <cassert>
+#include <chrono>
+#include <cmath>
+#include <cstdio>
+#include <iostream>
+#include <vector>
+
+constexpr size_t length = 1048576;
+std::vector<unsigned long long> population_vector, sample_vector,
+    original_vector;
+// std::vector<std::vector<unsigned long long>> result_storage;
+std::vector<bool> result_storage;
+
+long double sample_cdf(long double x) {
+  auto it = lower_bound(sample_vector.begin(), sample_vector.end(), x);
+  if (it == sample_vector.end()) {
+    return 1;
+  }
+  if (it == sample_vector.begin()) {
+    return 0;
+  }
+  auto it_prev = it - 1;
+  return ((long double)(it_prev - sample_vector.begin()) +
+          (long double)(x - *it_prev) / (*it - *it_prev)) /
+         (long double)(sample_vector.size() - 1);
+}
+
+inline long double safe_ceil(long double value) {
+  auto c = std::ceil(value);
+  return c == 0 ? 1 : c;
+}
+
+unsigned long long max_value = 0, min_value = 0xfffffffff;
+
+inline void store_into_vector(unsigned long long value) {
+  if (max_value < value) {
+    max_value = value;
+  }
+  if (min_value > value) {
+    min_value = value;
+  }
+  population_vector.push_back(value);
+}
+
+constexpr long sample_length = 1024;
+constexpr long test_size = 2048;
+
+void mian(long scale_size) {
+  const long split_size = test_size * scale_size;
+  result_storage.clear();
+  result_storage.resize(split_size, false);
+
+  const auto slice_size = 1.0 / (long double)(split_size);
+
+  for (long i = 0; i < test_size; i++) {
+    auto index = (int)safe_ceil(sample_cdf(original_vector[i + sample_length]) /
+                                slice_size) -
+                 1;
+    // printf("%Lf\n", index);
+    if (result_storage[index]) {
+      while (result_storage[++index]) {
+        assert(index < split_size);
+      }
+    }
+    result_storage[index] = true;
+  }
+}
+
+int main(int _argc, char const *_argv[]) {
+  assert((length & 1) == 0);
+  assert(length > 1024);
+
+  FILE *file = fopen("normal_distribution.txt", "r");
+  assert(file);
+  for (long long i; fscanf(file, "%lld ", &i) != EOF; store_into_vector(i))
+    ;
+  fclose(file);
+
+  assert(population_vector.size() == length);
+
+  original_vector = population_vector;
+
+  sample_vector = std::vector<unsigned long long>(
+      original_vector.begin(), original_vector.begin() + sample_length);
+  sample_vector.push_back(min_value);
+  sample_vector.push_back(max_value);
+
+  std::sort(sample_vector.begin(), sample_vector.end());
+
+  for (int i = 2; i <= 16; i++) {
+
+    auto start = std::chrono::high_resolution_clock::now();
+    mian(i);
+    auto end = std::chrono::high_resolution_clock::now();
+
+    auto element_size_max = 0;
+    for (auto element : result_storage) {
+      if (element_size_max < element) {
+        element_size_max = element;
+      }
+    }
+    assert(element_size_max == 1);
+    printf("scale: %d, time spend: %ldms\n", i,
+           duration_cast<std::chrono::microseconds>((end - start)).count());
+  }
+}
\ No newline at end of file