https://github.com/allanchain/namecross
Make a cross maze to include everyone's name!
https://github.com/allanchain/namecross
Last synced: 10 months ago
JSON representation
Make a cross maze to include everyone's name!
- Host: GitHub
- URL: https://github.com/allanchain/namecross
- Owner: AllanChain
- License: gpl-3.0
- Created: 2019-07-05T05:01:59.000Z (almost 7 years ago)
- Default Branch: master
- Last Pushed: 2019-07-10T03:25:02.000Z (almost 7 years ago)
- Last Synced: 2025-06-20T01:05:11.947Z (12 months ago)
- Language: Python
- Size: 2.25 MB
- Stars: 1
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# NameCross
Make a cross maze to include everyone's name! The connection made up a whole picture, you could not split it into two parts without spliting any name.
把所有人的名字组合成一个交叉迷宫！并且整个迷宫是一个整体，你不能把迷宫分成两部分而不分裂任何一个名字。
## Perfect Example
**The demos are outdated after a catastrophy (see commit 875352a)**
**下面的示例内容在大灾难后已经过时（见875352a）**
As far as I can see, the mazes below is perfect (No.2 and 3 is pruned by me):
我觉得下面这几个迷宫是完美的（图2和图3是我修过的）：
```
---wdlsry ----zgzss--- --slt-sswnp --xklsry---- lsry----pszl
--xjxym--- -d--mpszc--- -szcxjzqyth --sltzmh---- -wdlhrwhhzmh
cjkphrwswn -jxjhhzlycj- -dyjsjlycc- jxzdjcwdlx-- -nxycsydjcw-
szlscsyjx- -ckscsyjyjzq -jxycsyby-- sjlycsysqyz- --kfscyyzgz-
lyfscyylyc -sltcjklsry- scyjzyftyb- szyscyy-xcg- sjl-tcjkyth-
tybtdsszc- xymwdlyfwhh- tjmhgmwdl-- wqftybpjyjzy --szysxjyjzq
-tjyjzqg-- y-dythrwn--- xklhzmh---- nythrwhhmk-- --qyblyc-h--
whh-cyyz-- c-sqyb------ -lsry-h---- -b-----h---- --y--tm--h--
zmh---b--- ----b------- ---w-------
```
But there are also ugly ones:
但也有丑的：
```
----------x-
----------ks
---------slt
---------zx-
----xymsjlyf
wdlphrwhhycc
--szcsydjcjk
--rmscyythxj
sqyhzgzjyjhh
wyjzy---b---
nbz---------
--q---------
```
Where `-` represents an unused block. Note when you expand a character into a `1 * 1` box rather than `1 * 2`, it will be approximately a square.
其中`-`代表一个未被使用的空格。注意当你让一个字母占据`1 * 1`的格子而不是`1 * 2`时，它会大约变成一个正方形。
> Okey, I know there are two `cyy` and `sc`
>
> 好吧，我知道它有两个`sc`和`cyy`
Other solutions can be found at `solutions/`
你能在`solution/`文件夹中找到更多解
## Usage
```shell
usage: main.py [-h] --data DATA --output OUTPUT [-n N] [--use-colorama]
[--seed SEED | --random WxH]
optional arguments:
-h, --help show this help message and exit
--data DATA, -d DATA The name data file
--output OUTPUT, -o OUTPUT
Directory to store output
-n N Number of attempts
--use-colorama, -c Whether use colorama
--seed SEED, -s SEED The seed to apply
--random WxH, -r WxH Use random mode and specify its size, e.g. 12x12
# To run the example data set for 100 attempts and output to directory better/
python3 main.py --data data/TG2019303.txt --o better
# If you are using Windows cmd and has colorama installed, add a -c option
python3 main.py --data data/TG2019303.txt --o better -c
# To run 1000 attempts
python3 main.py --data data/TG2019303.txt --o better -n 1000
# To give it a seed file
python3 main.py --data data/TG2019303.txt --o better -s data/seed_one.txt
# To use random mode
python3 main.py --data data/TG2019303.txt --o better -r 11x11
```
```shell
usage: gif_maker.py [-h] --data DATA --input INPUT [--text TEXT] --output
OUTPUT
optional arguments:
-h, --help show this help message and exit
--data DATA, -d DATA Name data file
--input INPUT, -i INPUT
Input name map file
--text TEXT, -t TEXT Text to draw
--output OUTPUT, -o OUTPUT
Output directory
# To get the demo gif
gif_maker.py -d data/TG2019303.txt -i better/f465c4059c8ab543af0b0b842723ad13 -o demo.gif
# Or if you belong to another class
gif_maker.py -d data/TG2019319.txt -i better/ffffffffffffffffffffffffffffffff -t 319 -o demo.gif
```
## How It Works
It has **nothing to do with AI**, because it is just hard-coded evaluation with high weight of the often no used names, and a kiand of random choice. In general, 10 answers are found per thousand attempt. It takes 47s to complete 1000 attempts on my laptop.
它**和AI没有半毛钱关系** ， 因为它不过是预先写好的评估算法（加上对失败尝试时未被使用的名字的高权重）和随机选择。在我的笔记本上，1000次尝试会有10个解，要花47秒钟。
# ToDo
- [X] Add overall evaluation to get best output in all solutions
- [X] Different name to be the seed
- [ ] Better algorithm (maybe real ML)
- [X] Gif animation or HTML&JS interaction