[Bug] Fix generating nan values in tensor

- Gradient tensor values are inconsistently set to NaN
- NaN values caused incorrect backwarding in Neural Net
- Replacing malloc with calloc prevents memory allocation with value set to NaN

Signed-off-by: Donghyeon Jeong <djeong20@illinois.edu>

diff --git a/nntrainer/tensor/memory_pool.cpp b/nntrainer/tensor/memory_pool.cpp
index 5be93ea..782dce1 100644 (file)
--- a/nntrainer/tensor/memory_pool.cpp
+++ b/nntrainer/tensor/memory_pool.cpp
@@ -95,7 +95,7 @@ void MemoryPool::allocate() {
   if (mem_pool != nullptr)
     throw std::runtime_error("Memory pool is already allocated");
 
-  mem_pool = malloc(pool_size);
+  mem_pool = calloc(pool_size, 1);
   if (mem_pool == nullptr)
     throw std::runtime_error(
       "Failed to allocate memory: " + std::to_string(pool_size) + "bytes");