File size: 7,579 Bytes
07f0893
 
 
230eb64
c647402
230eb64
 
59ef627
 
 
 
230eb64
 
 
59ef627
230eb64
 
 
 
 
 
 
 
 
 
 
 
 
 
59ef627
230eb64
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
59ef627
230eb64
 
 
 
 
 
 
 
 
 
 
 
 
 
59ef627
230eb64
 
 
 
 
 
 
 
 
 
 
 
59ef627
230eb64
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
59ef627
693625c
230eb64
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
59ef627
693625c
230eb64
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
693625c
230eb64
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3b5724a
693625c
230eb64
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
---
license: mit
---

This repository contains the dataset for ["Breakeven complexity: A new perspective on neural partial differential equation solvers"](https://arxiv.org/abs/2605.15399).

Dataset detail:
- Navier-Stokes: Simulated via [Exponax](https://fkoehler.site/exponax/). Contains field "u". 20,000 training trajectories and 1,000 test trajectories. Shape (N, res, res, T).
- Kuramoto-Sivashinsky: Simulated via [Exponax](https://fkoehler.site/exponax/). Contains field "u". 20,000 training trajectories and 1,000 test trajectories. Shape (N, res, res, T).
- Gray-Scott: Simulated via [Exponax](https://fkoehler.site/exponax/). Contains field "u" and "v". 20,000 training trajectories seperated equally into 10 files, and 1,000 test trajectories. Shape (N, res, res, T).
- BreakFlow: Simulated via [PyFR](https://www.pyfr.org/). Contains field "velocity_x", ""velocity_y", "pressure", "vorticity", and "mask". 2,000 training trajectories and 50 test trajectories for each Reynolds bin. The combined folder is a shuffled collection of the data. The OOD(out of distribution) folder contrains train set of a shuffled collection of bin1 and bin2, and bin3 as test set. Shape (N, T, res, res).

Here are time matching statistics for the datasets:

# Navier-Stokes 
| Resolution | nRMSE | Time (seconds) |
| :--- | :--- | :--- |
| **10** | 0.710526 | 0.020756 |
| **20** | 0.274556 | 0.020724 |
| **30** | 0.079504 | 0.022734 |
| **40** | 0.036205 | 0.025200 |
| **50** | 0.018495 | 0.027116 |
| **60** | 0.009136 | 0.032884 |
| **70** | 0.005968 | 0.036457 |
| **80** | 0.003813 | 0.041083 |
| **90** | 0.002139 | 0.041915 |
| **100** | 0.001344 | 0.043408 |
| **256** | 0.000000 | 0.043537 |

# Kuramoto-Sivashinsky 1D
| Resolution | nRMSE | Time (seconds) |
| :--- | :--- | :--- |
| **8** | 0.840487 | 0.000209 |
| **10** | 0.694043 | 0.000236 |
| **12** | 0.579952 | 0.000225 |
| **14** | 0.579953 | 0.000234 |
| **16** | 0.576410 | 0.000200 |
| **18** | 0.590125 | 0.000230 |
| **20** | 0.590130 | 0.000318 |
| **22** | 0.516423 | 0.000243 |
| **24** | 0.430445 | 0.000232 |
| **26** | 0.430447 | 0.000254 |
| **28** | 0.320877 | 0.000243 |
| **30** | 0.188443 | 0.000238 |
| **32** | 0.188440 | 0.000237 |
| **34** | 0.076120 | 0.000240 |
| **36** | 0.046457 | 0.000212 |
| **38** | 0.046294 | 0.000243 |
| **40** | 0.023428 | 0.000240 |
| **42** | 0.011915 | 0.000250 |
| **44** | 0.011915 | 0.000243 |
| **46** | 0.006401 | 0.000256 |
| **48** | 0.003399 | 0.000243 |
| **50** | 0.003400 | 0.000251 |
| **52** | 0.001929 | 0.000369 |
| **54** | 0.001090 | 0.000250 |
| **56** | 0.001089 | 0.000250 |
| **58** | 0.000658 | 0.000290 |
| **60** | 0.000426 | 0.000262 |
| **62** | 0.000412 | 0.000373 |
| **64** | 0.000000 | 0.000310 |

# Kuramoto-Sivashinsky 2D
| Resolution | nRMSE | Time (seconds) |
| :--- | :--- | :--- |
| **54** | 0.504471 | 0.015612 |
| **60** | 0.217395 | 0.017008 |
| **64** | 0.148039 | 0.017242 |
| **70** | 0.070031 | 0.017442 |
| **74** | 0.045836 | 0.019862 |
| **80** | 0.015524 | 0.021907 |
| **84** | 0.004999 | 0.023360 |
| **90** | 0.001333 | 0.023940 |
| **94** | 0.000871 | 0.027558 |
| **100** | 0.000704 | 0.030900 |
| **256** | 0.000000 | 0.066734 |

# Kuramoto-Sivashinsky 3D
| Resolution | nRMSE | Time (seconds) |
| :--- | :--- | :--- |
| **48** | 0.439662 | 1.144816 |
| **50** | 0.444682 | 1.199737 |
| **52** | 0.171244 | 1.654107 |
| **54** | 0.073160 | 1.338130 |
| **56** | 0.074237 | 1.356794 |
| **58** | 0.031996 | 1.912379 |
| **60** | 0.013426 | 1.338568 |
| **62** | 0.012842 | 2.050640 |
| **64** | 0.000000 | 2.711006 |

# Gray-Scott
| Resolution | nRMSE | Time (seconds) |
| :--- | :--- | :--- |
| **10** | 0.803211 | 0.037729 |
| **20** | 0.722904 | 0.041896 |
| **30** | 0.635374 | 0.047774 |
| **40** | 0.587572 | 0.051580 |
| **50** | 0.546942 | 0.061180 |
| **60** | 0.453108 | 0.068281 |
| **70** | 0.326739 | 0.074282 |
| **80** | 0.245069 | 0.081521 |
| **90** | 0.180130 | 0.101329 |
| **100** | 0.129753 | 0.114296 |
| **110** | 0.091766 | 0.131567 |
| **120** | 0.064967 | 0.143684 |
| **130** | 0.045852 | 0.172376 |
| **140** | 0.034773 | 0.189323 |
| **150** | 0.024973 | 0.213734 |
| **160** | 0.016908 | 0.224823 |
| **170** | 0.015451 | 0.361601 |
| **180** | 0.014250 | 0.283169 |
| **190** | 0.010380 | 0.354393 |
| **200** | 0.003622 | 0.379086 |
| **210** | 0.002213 | 0.383786 |
| **220** | 0.002397 | 0.425519 |
| **230** | 0.002240 | 0.473417 |
| **256** | 0.000000 | 0.475097 |

# BreakFlow Combined
| Coarse size | nRMSE | Time (seconds) |
| :--- | :--- | :--- |
| **1.0000** | 0.000000 | 136.90 |
| **1.0443** | 0.128982 | 116.90 |
| **1.0905** | 0.129075 | 112.56 |
| **1.1388** | 0.130949 | 108.54 |
| **1.1892** | 0.131635 | 106.62 |
| **1.4142** | 0.148803 | 80.38 |
| **1.6818** | 0.161297 | 61.96 |
| **2.0000** | 0.183323 | 45.88 |
| **2.3784** | 0.185847 | 42.64 |
| **2.8284** | 0.188891 | 41.28 |
| **3.3636** | 0.222432 | 28.84 |
| **4.0000** | 0.225344 | 27.60 |
| **4.7568** | 0.240020 | 26.86 |
| **5.6569** | 0.240704 | 26.84 |
| **6.7272** | 0.271237 | 17.96 |
| **8.0000** | 0.276092 | 17.64 |
| **9.5137** | 0.288030 | 15.78 |
| **11.3137** | 0.292035 | 15.72 |
| **13.4543** | 0.307576 | 15.68 |
| **16.0000** | 0.310202 | 15.66 |
| **19.0274** | 0.331415 | 15.00 |

# BreakFlow Bin1
| Coarse size | nRMSE | Time (seconds) |
| :--- | :--- | :--- |
| **1.0000** | 0.000000 | 95.00 |
| **1.0443** | 0.002162 | 90.68 |
| **1.0905** | 0.002408 | 87.58 |
| **1.1388** | 0.002191 | 84.58 |
| **1.1892** | 0.002310 | 82.16 |
| **1.4142** | 0.002670 | 62.50 |
| **1.6818** | 0.004234 | 48.10 |
| **2.0000** | 0.006039 | 36.38 |
| **2.3784** | 0.007088 | 33.78 |
| **2.8284** | 0.007740 | 32.42 |
| **3.3636** | 0.017143 | 23.14 |
| **4.0000** | 0.018088 | 22.24 |
| **4.7568** | 0.019215 | 21.78 |
| **5.6569** | 0.022276 | 21.58 |
| **6.7272** | 0.042654 | 14.48 |
| **8.0000** | 0.048285 | 14.26 |
| **9.5137** | 0.063298 | 12.96 |
| **11.3137** | 0.072127 | 12.88 |
| **13.4543** | 0.092477 | 12.86 |
| **16.0000** | 0.093449 | 12.78 |

# BreakFlow Bin2
| Coarse size | nRMSE | Time (seconds) |
| :--- | :--- | :--- |
| **1.0000** | 0.000000 | 99.24 |
| **1.0443** | 0.054530 | 96.30 |
| **1.0905** | 0.058209 | 92.84 |
| **1.1388** | 0.048698 | 89.30 |
| **1.1892** | 0.055428 | 87.10 |
| **1.4142** | 0.081105 | 66.50 |
| **1.6818** | 0.088567 | 51.16 |
| **2.0000** | 0.123205 | 39.36 |
| **2.3784** | 0.126006 | 36.22 |
| **2.8284** | 0.125205 | 35.32 |
| **3.3636** | 0.179135 | 24.54 |
| **4.0000** | 0.179902 | 24.02 |
| **4.7568** | 0.199868 | 23.18 |
| **5.6569** | 0.211182 | 23.32 |
| **6.7272** | 0.250170 | 16.12 |
| **8.0000** | 0.255241 | 15.60 |
| **9.5137** | 0.285511 | 13.48 |
| **11.3137** | 0.278287 | 13.66 |
| **13.4543** | 0.314424 | 13.52 |
| **16.0000** | 0.314749 | 13.80 |

# BreakFlow Bin3
| Coarse size | nRMSE | Time (seconds) |
| :--- | :--- | :--- |
| **1.0000** | 0.000000 | 167.48 |
| **1.0443** | 0.330253 | 163.74 |
| **1.0905** | 0.332230 | 157.28 |
| **1.1388** | 0.336337 | 151.76 |
| **1.1892** | 0.337167 | 150.62 |
| **1.4142** | 0.362633 | 112.16 |
| **1.6818** | 0.391089 | 86.62 |
| **2.0000** | 0.420725 | 61.88 |
| **2.3784** | 0.424446 | 57.90 |
| **2.8284** | 0.433728 | 56.10 |
| **3.3636** | 0.471017 | 38.82 |
| **4.0000** | 0.478041 | 36.52 |
| **4.7568** | 0.500976 | 35.62 |
| **5.6569** | 0.488654 | 35.62 |
| **6.7272** | 0.520888 | 23.28 |
| **8.0000** | 0.524751 | 23.08 |
| **9.5137** | 0.515281 | 20.32 |
| **11.3137** | 0.525692 | 20.64 |
| **13.4543** | 0.515828 | 20.64 |
| **16.0000** | 0.522407 | 20.40 |