Donghak PARK

FullyConnected Layer Reordering for TF Lite Exporter

5

When Export ```NNTrainer``` to ```Tensorflow Lite``` we need to reordering --> Implement based on #1892 - [x] Implement Flatten Reordering - [x] Test various FC Case - [x] Make Unit...

automated memory & latency test

1

Can this application be built along with gbs and/or pdebuild so that "nntrainer-example-alexnet.rpm/deb" can be published ? If that's possible, you can generate an automated memory & latency test easily:...

[TF Exporter] FullyConnected Layer Weights Transpose for TFLite Export @open sesame 10/21 18:26

10

This PR for TFlite export Transpose FullyConnected Layer's weights (NCHW --> NHWC) in shape 1:3:2:4 & FC with unit 1 Before reordering 0 1 2 3 4 5 6 7...

[typo] Fix typo error @open sesame 10/21 11:43

1

Fix the typo errors in bellow dir - Applications/ - nntrainer/optimizers/ Signed-off-by: DonghakPark

[fix] fix coverity_tizen issue @open sesame 10/21 09:03

1

- Add try, catch statement to catch exception - Add NNTR_THROW_IF to check return Signed-off-by: DonghakPark

[TF Lite Export] "supportInPlace = True" Error while export

1

when ```supportInPlace``` is ```true``` Flatten Layer make nullptr Error and some other layers also make similar error

Add Depthwise 2D Convolution Layer

1

This layer is necessary to support various applications such as SV. - Depthwise convolution is a type of convolution in which each input channel is convolved with a different kernel...

[Wait for #2568][Mixed] Mixed Precision Layer update

4

This PR is to update the mixed precision layer. - integrate #2568 & #2455 - will update more test **Self evaluation:** 1. Build test: [X]Passed [ ]Failed [ ]Skipped 2....

[Application] Add Mixed precision Application

1

Add Mixed Precision example on Application - this example can guide developer to handle fp16 example - we can test & eval our model end-to-end - we can optimize base...

Support Convolution&Batchnorm Fusing for Optimized Inference Mode

3

One of the ways to accelerate running the NNTrainer in inference mode is to fuse operations. We are currently using this fusion when exporting to TensorFlow Lite. By applying it...