Hugging Face's logo

Spaces:
Hanxiaofeng123
/
Deepcube
Sleeping

App Files Community
Deepcube
File size: 4,984 Bytes 
b570cf2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
 
<!DOCTYPE html>
<html>
  <head>

    <title>DeepCubeA</title>
    <meta name="viewport" content="width=device-width, initial-scale=1.0">

    <!-- Bootstrap -->
    <link href="//cdnjs.cloudflare.com/ajax/libs/twitter-bootstrap/3.3.7/css/bootstrap.min.css" rel="stylesheet">
    <link rel="stylesheet" href="/static/main.css">

    <!--[if lt IE 9]>
          <script src="//cdnjs.cloudflare.com/ajax/libs/html5shiv/3.7.3/html5shiv.min.js"></script>
          <script src="//cdnjs.cloudflare.com/ajax/libs/respond.js/1.4.2/respond.min.js"></script>
    <![endif]-->

    <!-- 移除重复的引入代码 -->
    <link href="//cdnjs.cloudflare.com/ajax/libs/twitter-bootstrap/3.3.7/css/bootstrap.min.css" rel="stylesheet">
    <link rel="stylesheet" href="/static/main.css">


<link rel="shortcut icon" href="static/favicon.ico">
<meta name="twitter:card" content="summary_large_image">
<meta name="twitter:title" content="DeepCubeA: Solve the Rubik's cube With Deep Learning">
<meta name="twitter:description" content="DeepCubeA trains itself using deep learning and reinforcement learning. It then uses A* search to solve the cube.">
<meta name="twitter:image" content="https://www.ics.uci.edu/~fagostin/cube.jpg">


  </head>
  <body>
    
    
<div class="container">
  <center>
  <h1>Solve the Rubik's Cube Using Deep Learning</h1>
	<div id="program_container" class="container">
		<div id="solution_container" class="container">
			<button disabled id="first_state" type="button">&#060;&#060;</button>
			<button disabled id="prev_state" type="button">&#060;</button>
			<button disabled id="next_state" type="button">&#062;</button>
			<button disabled id="last_state" type="button">&#062;&#062;</button>
			<h4 id="solution_text">Solution:</h4>
		</div>

		<div id="cube_div" class="container" style="padding: 60px">
			<section class="cube_container">
				<div id="cube">
					<figure id="front" class="front">
					</figure>
					<figure id="back" class="back">
					</figure>
					<figure id="right" class="right">
					</figure>
					<figure id="left" class="left">
					</figure>
					<figure id="top" class="top">
					</figure>
					<figure id="bottom" class="bottom">
					</figure>
				</div>
			</section>
		</div>

		<br>

		<div id="button_container" class="container">
			<button disabled id="scramble" type="button">Scramble</button>
			<button disabled id="solve" type="button">Solve!</button>
			<button disabled id="clear" type="button">Clear</button>
		</div>
	</div>
  </center>
</div>

<br>
<center>
	<h4>
		Use the mouse to turn the cube. &#13;&#10;
	</h4>
	<h4>
		Turn the faces with the U/D/L/R/B/F keys.
		Hold shift to turn faces couter-clockwise.
	</h4>
	<h4>
		Press scramble to randomly scramble the cube.
		Press solve to solve the cube using deep learning!
	</h4>
	<h5>
		<i>Note: Due to model capacity limitations, there might be cases where the model cannot solve in reasonable time.</i>  
	</h5>
</center>

<br>
<div class="well well-lg">
	<center>
	<a href="https://github.com/forestagostinelli/DeepCubeA">Code (Latest)</a> - <a href="https://github.com/xiaofeng218/DeepcubeA">Code (Reproduce)</a>
	<br>
	<br>
	<span style='color:black'>If you find this research useful to your work, please cite the following papers:</span><br><br>

	<span class="paper-title"><a href="static/files/SolvingTheRubiksCubeWithDeepReinforcementLearningAndSearch_Final.pdf">Solving the Rubik's Cube with Deep Reinforcement Learning and Search</a></span><br>
	<!--span class="paper-title"><a href="https://www.nature.com/articles/s42256-019-0070-z.epdf?shared_access_token=-pCSsZa_J9bM8VyXLZLRctRgN0jAjWel9jnR3ZoTv0Osb8UCgUm5AQaSCMHWqWzsyV3KBcb13SAW-9IL1pAGd1HcSk40JSEjhoaBAi0ePvYh_5Dul6LvK0oJY1KI0ULo9O9HCut_y7aCTc93Th8m5g%3D%3D">Solving the Rubik's Cube with Deep Reinforcement Learning and Search</a></span><br-->
	<span class="paper-authors">Forest Agostinelli*, Stephen McAleer*, Alexander Shmakov*, Pierre Baldi</span><br>
	<span class="paper-journal"><i>Nature Machine Intelligence</i>, Volume 1</span>,
	<span class="paper-year">2019</span>
	<br>
	<br>
	<span class="paper-title"><a href="https://openreview.net/pdf?id=Hyfn2jCcKm">Solving the Rubik's Cube with Approximate Policy Iteration</a></span><br>
	<span class="paper-authors">Stephen McAleer*, Forest Agostinelli*, Alexander Shmakov*, Pierre Baldi</span><br>
	<span class="paper-journal"><i>In Proceedings of the 7th International Conference on Learning Representations (ICLR)</i></span>,
	<span class="paper-year">2019</span>
	</center>
</div>

<center>
	<h5>
		Note: For ease of maintenance, this updated version solves the cube using CPUs instead of a GPU.
		Therefore, solve times will be around 20 seconds instead of 1 second.
	</h5>
</center>

<script src="https://ajax.googleapis.com/ajax/libs/jquery/3.2.1/jquery.min.js"></script>
<script src="/static/main.js"></script>
<script src="//cdnjs.cloudflare.com/ajax/libs/twitter-bootstrap/3.3.7/js/bootstrap.min.js"></script>

  </body>
</html>