thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-21 15:12:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #16135 from hexiaoting/dev_rankCorr

Browse Source 
Fix inconsistent behaviour of rankCorr function
This commit is contained in:
alexey-milovidov 2020-10-22 01:15:50 +03:00 committed by GitHub
commit cfaf82ba0c
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 13 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
src/AggregateFunctions/AggregateFunctionRankCorrelation.h
View File

@ -21,10 +21,6 @@
#include <type_traits>
namespace ErrorCodes
{
extern const int BAD_ARGUMENTS;
}
namespace DB
{
@ -138,23 +134,18 @@ public:
const auto & value = this->data(place).values;
size_t size = this->data(place).size_x;
if (size < 2)
{
throw Exception("Aggregate function " + getName() + " requires samples to be of size > 1", ErrorCodes::BAD_ARGUMENTS);
}
//create a copy of values not to format data
// create a copy of values not to format data
PODArrayWithStackMemory<std::pair<Float64, Float64>, 32> tmp_values;
tmp_values.resize(size);
for (size_t j = 0; j < size; ++ j)
tmp_values[j] = static_cast<std::pair<Float64, Float64>>(value[j]);
//sort x_values
// sort x_values
std::sort(std::begin(tmp_values), std::end(tmp_values), ComparePairFirst<std::greater>{});
for (size_t j = 0; j < size;)
{
//replace x_values with their ranks
// replace x_values with their ranks
size_t rank = j + 1;
size_t same = 1;
size_t cur_sum = rank;
@ -166,9 +157,9 @@ public:
{
// rank of (j + 1)th number
rank += 1;
same++;
++same;
cur_sum += rank;
j++;
++j;
}
else
break;
@ -178,16 +169,16 @@ public:
Float64 insert_rank = static_cast<Float64>(cur_sum) / same;
for (size_t i = cur_start; i <= j; ++i)
tmp_values[i].first = insert_rank;
j++;
++j;
}
//sort y_values
// sort y_values
std::sort(std::begin(tmp_values), std::end(tmp_values), ComparePairSecond<std::greater>{});
//replace y_values with their ranks
// replace y_values with their ranks
for (size_t j = 0; j < size;)
{
//replace x_values with their ranks
// replace x_values with their ranks
size_t rank = j + 1;
size_t same = 1;
size_t cur_sum = rank;
@ -199,9 +190,9 @@ public:
{
// rank of (j + 1)th number
rank += 1;
same++;
++same;
cur_sum += rank;
j++;
++j;
}
else
{
@ -213,10 +204,10 @@ public:
Float64 insert_rank = static_cast<Float64>(cur_sum) / same;
for (size_t i = cur_start; i <= j; ++i)
tmp_values[i].second = insert_rank;
j++;
++j;
}
//count d^2 sum
// count d^2 sum
Float64 answer = static_cast<Float64>(0);
for (size_t j = 0; j < size; ++ j)
answer += (tmp_values[j].first - tmp_values[j].second) * (tmp_values[j].first - tmp_values[j].second);