NVIDIA/nvbench
1
0
Fork 0
mirror of https://github.com/NVIDIA/nvbench.git synced 2026-04-20 06:48:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

Encode generic numerics as strings in JSON format.

Browse Source 
JSON encodes all numerics as `double`s, which would truncate int64_t.
Write all generic numbers from `named_values` objects as strings.
These are still distinguishable by their `type` properties.
This commit is contained in:
Allison Vacanti
2021-06-22 16:10:52 -04:00
parent f1715786fd
commit 668a98ef16
1 changed files with 7 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
nvbench/json_printer.cu
View File

@@ -24,6 +24,8 @@
#include <nvbench/device_manager.cuh>
#include <nvbench/summary.cuh>
#include <fmt/format.h>
#include <nlohmann/json.hpp>
#include <cstdint>
@@ -47,12 +49,15 @@ void write_named_values(JsonNode &node, const nvbench::named_values &values)
{
case nvbench::named_values::type::int64:
value["type"] = "int64";
value["value"] = values.get_int64(value_name);
// Write as a string; JSON encodes all numbers as double-precision
// floats, which would truncate int64s.
value["value"] = fmt::to_string(values.get_int64(value_name));
break;
case nvbench::named_values::type::float64:
value["type"] = "float64";
value["value"] = values.get_float64(value_name);
// Write as a string for consistency with int64.
value["value"] = fmt::to_string(values.get_float64(value_name));
break;
case nvbench::named_values::type::string: