yoshitomo-matsubara
commited on
Commit
•
3ffc7ce
1
Parent(s):
9cd8aba
added formula table
Browse files
README.md
CHANGED
@@ -49,16 +49,59 @@ task_ids: []
|
|
49 |
## Dataset Description
|
50 |
|
51 |
- **Homepage:**
|
52 |
-
- **Repository:**
|
53 |
- **Paper:** Rethinking Symbolic Regression Datasets and Benchmarks for Scientific Discovery
|
54 |
-
- **Point of Contact:**
|
55 |
|
56 |
### Dataset Summary
|
57 |
|
58 |
Our SRSD (Feynman) datasets are designed to discuss the performance of Symbolic Regression for Scientific Discovery.
|
59 |
We carefully reviewed the properties of each formula and its variables in [the Feynman Symbolic Regression Database](https://space.mit.edu/home/tegmark/aifeynman.html) to design reasonably realistic sampling range of values so that our SRSD datasets can be used for evaluating the potential of SRSD such as whether or not a SR method con (re)discover physical laws from such datasets.
|
60 |
|
61 |
-
This is the Medium set of our SRSD-Feynman datasets, which consists of 40 different physics formulas
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
62 |
|
63 |
|
64 |
### Supported Tasks and Leaderboards
|
@@ -83,7 +126,7 @@ For each dataset, we have
|
|
83 |
1. train split (txt file, whitespace as a delimiter)
|
84 |
2. val split (txt file, whitespace as a delimiter)
|
85 |
3. test split (txt file, whitespace as a delimiter)
|
86 |
-
4. true equation (pickle file)
|
87 |
|
88 |
### Data Splits
|
89 |
|
|
|
49 |
## Dataset Description
|
50 |
|
51 |
- **Homepage:**
|
52 |
+
- **Repository:** https://github.com/omron-sinicx/srsd-benchmark
|
53 |
- **Paper:** Rethinking Symbolic Regression Datasets and Benchmarks for Scientific Discovery
|
54 |
+
- **Point of Contact:** [Yoshitomo Matsubara](mailto:yoshitom@uci.edu) [Yoshitaka Ushiku](mailto:yoshitaka.ushiku@sinicx.com)
|
55 |
|
56 |
### Dataset Summary
|
57 |
|
58 |
Our SRSD (Feynman) datasets are designed to discuss the performance of Symbolic Regression for Scientific Discovery.
|
59 |
We carefully reviewed the properties of each formula and its variables in [the Feynman Symbolic Regression Database](https://space.mit.edu/home/tegmark/aifeynman.html) to design reasonably realistic sampling range of values so that our SRSD datasets can be used for evaluating the potential of SRSD such as whether or not a SR method con (re)discover physical laws from such datasets.
|
60 |
|
61 |
+
This is the Medium set of our SRSD-Feynman datasets, which consists of the following 40 different physics formulas:
|
62 |
+
|
63 |
+
| ID | Formula |
|
64 |
+
|-----------|---------------------------------------------------------------------------------------------|
|
65 |
+
| I.8.14 | \\(d = \sqrt{(x_2-x_1)^2+(y_2-y_1)^2}\\) |
|
66 |
+
| I.10.7 | \\(m = \frac{m_0}{\sqrt{1-\frac{v^2}{c^2}}}\\) |
|
67 |
+
| I.11.19 | \\(A = x_1 y_1+x_2 y_2+x_3 y_3\\) |
|
68 |
+
| I.12.2 | \\(F = \frac{q_1 q_2}{4 \pi \epsilon r^2}\\) |
|
69 |
+
| I.12.11 | \\(F = q \left(E + B v \sin\left(\theta\right)\right)\\) |
|
70 |
+
| I.13.4 | \\(K = \frac{1}{2} m (v^2 + u^2 + w^2)\\) |
|
71 |
+
| I.13.12 | \\(U = G m_1 m_2 \left(\frac{1}{r_2}-\frac{1}{r_1}\right)\\) |
|
72 |
+
| I.15.10 | \\(p = \frac{m_0 v}{\sqrt{1-v^2/c^2}}\\) |
|
73 |
+
| I.16.6 | \\(v_1 = \frac{u+v}{1+u v/c^2}\\) |
|
74 |
+
| I.18.4 | \\(r = \frac{m_1 r_1+m_2 r_2}{m_1+m_2}\\) |
|
75 |
+
| I.24.6 | \\(E = \frac{1}{4} m (\omega^2+\omega_0^2) x^2\\) |
|
76 |
+
| I.29.4 | \\(k = \frac{\omega}{c}\\) |
|
77 |
+
| I.32.5 | \\(P = \frac{q^2 a^2}{6 \pi \epsilon c^3}\\) |
|
78 |
+
| I.34.8 | \\(\omega = \frac{q v B}{p}\\) |
|
79 |
+
| I.34.10 | \\(\omega = \frac{\omega_0}{1-v/c}\\) |
|
80 |
+
| I.34.27 | \\(W = \frac{h}{2 \pi} \omega\\) |
|
81 |
+
| I.38.12 | \\(r = 4 \pi \epsilon \frac{\left(h/\left(2 \pi\right)\right)^2}{m q^2}\\) |
|
82 |
+
| I.39.10 | \\(U = \frac{3}{2} P V\\) |
|
83 |
+
| I.39.11 | \\(U = \frac{P V}{\gamma-1}\\) |
|
84 |
+
| I.43.31 | \\(D = \mu k T\\) |
|
85 |
+
| I.43.43 | \\(\kappa = \frac{1}{\gamma - 1} \frac{k v}{\sigma_c}\\) |
|
86 |
+
| I.48.2 | \\(E = \frac{m c^2}{\sqrt{1-v^2/c^2}}\\) |
|
87 |
+
| II.6.11 | \\(\phi = \frac{1}{4 \pi \epsilon} \frac{p \cos\theta}{r^2}\\) |
|
88 |
+
| II.8.7 | \\(U = \frac{3}{5} \frac{Q^2}{4 \pi \epsilon a}\\) |
|
89 |
+
| II.11.3 | \\(x = \frac{q E}{m (\omega_0^2-\omega^2)}\\) |
|
90 |
+
| II.21.32 | \\(\phi = \frac{q}{4 \pi \epsilon r (1-v/c)}\\) |
|
91 |
+
| II.34.2 | \\(\mu = \frac{q v r}{2}\\) |
|
92 |
+
| II.34.2a | \\(I = \frac{q v}{2 \pi r}\\) |
|
93 |
+
| II.34.29a | \\(\mu = \frac{q h}{4 \pi m}\\) |
|
94 |
+
| II.37.1 | \\(E = \mu (1+\chi) B\\) |
|
95 |
+
| III.4.32 | \\(n = \frac{1}{\exp(h \omega/2 \pi k T) - 1}\\) |
|
96 |
+
| III.8.54 | \\(|C|^2 = \sin^2 \frac{2 \pi A t}{h}\\) |
|
97 |
+
| III.13.18 | \\(v = \frac{4 \pi A b^2}{h} k\\) |
|
98 |
+
| III.14.14 | \\(I = I_0 \left(\exp\left(q \Delta V/\kappa T\right)-1\right)\\) |
|
99 |
+
| III.15.12 | \\(E = 2 A (1-\cos k d)\\) |
|
100 |
+
| III.15.14 | \\(m = \frac{h^2}{8 \pi^2 A b^2}\\) |
|
101 |
+
| III.17.37 | \\(f = \beta (1+\alpha \cos\theta)\\) |
|
102 |
+
| III.19.51 | \\(E = -\frac{m q^4}{2 (4 \pi \epsilon)^2 (h/(2 \pi))^2 n^2}\\) |
|
103 |
+
| B8 | \\(U = \frac{E}{1+\frac{E}{m c^2} (1-\cos\theta)}\\) |
|
104 |
+
| B18 | \\(\rho = \frac{3}{8 \pi G} \left(\frac{c^2 k_\text{f}}{a_\text{f}^2}+H^2\right)\\) |
|
105 |
|
106 |
|
107 |
### Supported Tasks and Leaderboards
|
|
|
126 |
1. train split (txt file, whitespace as a delimiter)
|
127 |
2. val split (txt file, whitespace as a delimiter)
|
128 |
3. test split (txt file, whitespace as a delimiter)
|
129 |
+
4. true equation (pickle file for sympy object)
|
130 |
|
131 |
### Data Splits
|
132 |
|