projects / platform / core / ml / nntrainer.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: ad5e451) | patch
[DEBUG] Add interface for getting execution order
authorJiho Chu <jiho.chu@samsung.com>
Mon, 13 Feb 2023 09:32:08 +0000 (18:32 +0900)
committerjijoong.moon <jijoong.moon@samsung.com>
Tue, 4 Apr 2023 01:39:05 +0000 (10:39 +0900)
commitbb5e4c42d8a05e234a44c1dbd351e27a54b740ca
treef4b7289a5613a14b70dd9e2471d0d6f1ac1d44e1tree | snapshot
parentad5e4512268db350d3ab3753d43e6c5d38f6d4decommit | diff
[DEBUG] Add interface for getting execution order

This patch implemnts debugging feature for execution order.

Execution order for each tensor is decided while initalize and finalize
phase. To check the validity of them, getting execution order interface
is added to network graph class. It can gather both variable and
gradient of the tensors.

Signed-off-by: Jiho Chu <jiho.chu@samsung.com>
nntrainer/graph/network_graph.cpp diff | blob | history
nntrainer/graph/network_graph.h diff | blob | history
nntrainer/layers/layer_node.cpp diff | blob | history
nntrainer/layers/layer_node.h diff | blob | history
nntrainer/tensor/manager.cpp diff | blob | history
nntrainer/tensor/manager.h diff | blob | history
Domain: Machine Learning / ML Framework;