git
make
clang
g++
python >= 3.5
setuptools
torch >= 1.0
This will be installed during installation process.
Make sure you have installed all the requirements before installing.
# If 'position-learning' and 'position_learning' directory exists at project root,
# erase them before install.
make install
# run for very small size code.
make fasttest
# run for little bigger code than above one.
make demo
You can see the example at demodata/TC_1204B/input
and demodata/TC_1204B/output
.
It should be strictly followed.
$ tree demodata/TC_1204B/input
demodata/TC_1204B/input
├── 1
├── 10
├── 11
├── 12
├── 13
├── 14
├── 2
├── 3
├── 4
├── 5
├── 6
├── 7
├── 8
└── 9
0 directories, 14 files
$ tree demodata/TC_1204B/output
demodata/TC_1204B/output
├── 1
├── 10
├── 11
├── 12
├── 13
├── 14
├── 2
├── 3
├── 4
├── 5
├── 6
├── 7
├── 8
└── 9
0 directories, 14 files
As you can see, the corresponding output of the input file input/1
is output/1
. You can configure related setting at src/setting.py
's TESTCASE_STARTNUM
, TESTCASE_PREFIX
, and TESTCASE_POSTFIX
.
python3 app.py [Source File] [Input Testcase Directory] [Output Testcase Directory] [Result Output Json Filename]
If you turn on the JSONFILE_OUT
flag in setting.py
, you should put the argument Result Output Json Filename
. If JSONFILE_OUT
flag turn off, the argument Result Output Json Filename
will be ignored.