lengjia / nas_npu

for train the nas_npu

nas for processor-aware by BO and OPT

This is a python implemention of searching toward pareto-optimal processor-aware network architecture search by BO and optimal transport.

This repo in based on 'NASBOT', and is changed for processor aware nureal architecture search. by this way, the opensource neural network accelerator Eyeriss is use for hardware test. we mainly focus on the accuacy and energy consumpation.

For more details, please see our paper below.

For questions and bug reports please email [email protected].

Installation

• Download the package.

$ git clone https://github.com/kirthevasank/nasbot.git

• Install the following packages packages via pip: cython, POT (Python Optimal Transport), graphviz and pygraphviz. graphviz and pygraphviz are only needed to visualise the networks and are not necessary to run nasbot. However, some unit tests may fail.

$ pip install cython POT graphviz pygraphviz

In addition to the above, you will need numpy and scipy which can also be pip installed.

• Now set HOME_PATH in the set_up file to the parent directory of nasbot, i.e. HOME_PATH=<path/to/parent/directory>/nasbot. Then source the set up file.

$ source set_up

• Next, you need to build the direct fortran library. For this cd into utils/direct_fortran and run bash make_direct.sh. You will need a fortran compiler such as gnu95. Once this is done, you can run python simple_direct_test.py to make sure that it was installed correctly. The default version of NASBOT can be run without direct, but some unit tests might fail.

• Finally, you need to install tensorflow to execute the MLP/CNN demos on GPUs.

$ pip install tensorflow-gpu

Testing the Installation: To test the installation, run bash run_all_tests.sh. Some of the tests are probabilistic and could fail at times. If this happens, run the same test several times and make sure it is not consistently failing. Running all tests will take a while. You can run each unit test individually simpy via python unittest_xxx.py.

Getting started

License

This software is released under the MIT license. For more details, please refer [LICENSE.txt].

"Copyright 2018 [email protected]"

• For questions and bug reports please email [email protected]