+--------+---------------+----------------------------+---------------------------------------+---------------------+
| 1.0 | 2018.06.22 | Final DR1 version | Vostokov Sergey | Sung-Jae Lee |
+--------+---------------+----------------------------+---------------------------------------+---------------------+
-| 2.0 | 2018.09.03 | DR2 version | Roman Rusyaev | Sung-Jae Lee |
+| 2.0 | 2018.09.03 | Add ModelIR,Caffe frontend,| Vostokov Sergey | Sung-Jae Lee |
+| | | CPU Soft backend, CLI | | |
+--------+---------------+----------------------------+---------------------------------------+---------------------+
|
- Support ARM GPU (Mali)
- Support Tizen OS
- Support SmartMachine OS (Optional)
-
|
.. list-table:: Table 1-1. Target Model
* At the time of writing Caffe **does not** have a DepthwiseConv layer.
* `Split` layer, quite surprisingly, just makes a few identical copies of the input (bottom) blob; it doesn't actually split the blob into parts.
-
{Generate the code} Detailed Design
-----------------------------------
data offset is defined by ``Serializer``.
+Generation of C++ source code for GPU
+#####################################
+
+This backend has to generate the source code in C++ language to perform NN inference using GPU device. This feature of NN Compiler is under developement now. The detailed design will be provided till the project completion.
+
Interface Design
================