projects / platform / core / ml / nntrainer.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: ca3e4fa) | patch
Fix potential bug in pooling 2D layer
authorJihoon Lee <jhoon.it.lee@samsung.com>
Tue, 7 Jul 2020 02:44:18 +0000 (11:44 +0900)
committerJijoong Moon <jijoong.moon@samsung.com>
Tue, 7 Jul 2020 03:56:34 +0000 (12:56 +0900)
commitb8a55d2fd6bc8fab07c5cbd4ffe3d61107ab4b56
treeae91b56b51194026329cea8f76acc58e3c1b4b97tree | snapshot
parentca3e4fa3528d36a45f87507cc91557c0a5fab434commit | diff
Fix potential bug in pooling 2D layer

When pooling2d_layer is initialized once as a default.
and initialized again as a `global_max` | `global_pooling`

`output.width`, `output.height` is set to 2.

Although following scenario is highly unlikely, initializing layer
should be done deterministically

**Self evaluation:**
1. Build test: [X]Passed [ ]Failed [ ]Skipped
2. Run test: [X]Passed [ ]Failed [ ]Skipped

Signed-off-by: Jihoon Lee <jhoon.it.lee@samsung.com>
nntrainer/src/pooling2d_layer.cpp diff | blob | history
Domain: Machine Learning / ML Framework;