[Fix] handle edge case for countNonTrainable
authorJihoon Lee <jhoon.it.lee@samsung.com>
Mon, 26 Apr 2021 08:59:55 +0000 (17:59 +0900)
committerJijoong Moon <jijoong.moon@samsung.com>
Mon, 3 May 2021 00:40:15 +0000 (09:40 +0900)
commitb92ac856913186d63d58b75441588befa7707a36
tree26b2f58de9cabc9e9cfbe2137a63ce4cc7d8dc2d
parent0588cabc8d9903838ba12a387d5cc8776e29cf5a
[Fix] handle edge case for countNonTrainable

**Changes proposed in this PR:**
- Handle edge case when every layer is non trainable
for `countNonTrainableLayersAtBegin`
- Trivial code cleans on simpleshot task_runner / neuralnetwork

**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>
Applications/SimpleShot/meson.build
Applications/SimpleShot/task_runner.cpp
nntrainer/graph/network_graph.cpp
nntrainer/models/neuralnet.cpp