Catalog dataframe¶
The first things we need to know about the data are
- which simulations are available, and
- what physical systems those simulations represent.
We provide a pre-built interface to all of the data at https://data.black-holes.org/. That page actually runs Python directly in the browser, and calls a subset of the code described in this tutorial — though with nice interactive tables and plots.
Here, we explain some of the fundamentals, and give examples of how to
use the data directly. To begin, we load the dataframe object with
the sxs.load function:
In [ ]:
Copied!
import sxs
df = sxs.load("dataframe", tag="3.0.0")
import sxs
df = sxs.load("dataframe", tag="3.0.0")
Note the argument tag, which specifies the version of the data that you want to load. If you do not specify this argument, sxs will search for the most recent tag on github, and use that. However, to ensure that you can reproduce your own results, it is always a good idea to specify the tag explicitly. More importantly, when you publish results using data from the SXS catalog, you should cite the tag you used so that others can also reproduce your work.
The first time you call this function, or the first time you ask for a new tag, it will attempt to download the latest data from github and cache it locally, as described in the previous notebook.
The returned object is a pandas.DataFrame, which is the standard way to represent tabular data in Python. The index of the dataframe is the SXS ID, which is a unique identifier for each simulation. The SXS ID is a string of the form SXS:BBH:1234, where BBH indicates that this is a binary black hole simulation, and 1234 is a unique number assigned to this simulation. The SXS ID can be used to look up the simulation in the SXS catalog. The dataframe also has a number of columns that describe the simulation, including the physical parameters of the system, as well as other important metadata:
In [2]:
Copied!
df
df
Out[2]:
| deprecated | reference_mass_ratio | reference_chi_eff | reference_chi1_perp | reference_chi2_perp | reference_eccentricity | reference_eccentricity_bound | reference_time | reference_dimensionless_spin1_x | reference_dimensionless_spin1_y | ... | initial_position2 | number_of_orbits | number_of_orbits_from_start | number_of_orbits_from_reference_time | DOI_versions | keywords | date_link_earliest | date_run_earliest | date_run_latest | date_postprocessing | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SXS:BBH:0001 | True | 1.000000 | 1.215639e-07 | 9.733309e-10 | 1.429939e-09 | 0.000257 | 0.000257 | 320.000000 | 7.097521e-10 | -6.660518e-10 | ... | [-8.999999874197535, -2.86646421e-08, 0.0] | 28.119083 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v1.6, v2.0] | [Non-Spinning, ManyMergers-QuasiCircular, depr... | 2012-07-31 17:17:11 | 2012-08-01 01:46:59 | 2012-08-17 20:48:06 | 2019-03-30 13:41:50 |
| SXS:BBH:0002 | True | 1.000000 | 9.399402e-08 | 7.181763e-10 | 1.473155e-09 | 0.000175 | 0.000175 | 310.000000 | 6.507563e-10 | -3.037983e-10 | ... | [-9.49999971865619, 2.0830027e-07, 0.0] | 32.421068 | NaN | NaN | [, v1.1, v1.2, v1.4, v1.5, v1.6, v1.7, v2.0] | [Non-Spinning, ManyMergers-QuasiCircular, depr... | 2012-07-31 17:17:11 | 2012-08-01 02:24:21 | 2012-08-26 06:41:57 | 2019-03-30 13:41:50 |
| SXS:BBH:0003 | True | 1.000440 | 2.707179e-04 | 4.993912e-01 | 7.428880e-08 | 0.000287 | 0.000287 | 210.000000 | 4.965977e-01 | 5.274908e-02 | ... | [-9.507365032939443, -0.0023408782825215, 0.0] | 32.337574 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0] | [Precessing-Spins, ManyMergers-QuasiCircular, ... | 2012-08-07 19:02:51 | 2012-08-18 20:24:58 | 2012-08-22 10:48:06 | 2019-03-30 13:41:50 |
| SXS:BBH:0004 | True | 1.000459 | -2.498083e-01 | 5.018106e-11 | 1.402755e-09 | 0.000380 | 0.000380 | 640.000000 | -2.424428e-12 | -1.449169e-11 | ... | [-9.50286409695725, 0.0031879777731917, 0.0] | 30.188749 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0] | [Aligned-Spins, ManyMergers-QuasiCircular, dep... | 2012-08-07 19:02:51 | 2012-08-18 20:24:58 | 2012-08-25 10:38:13 | 2019-03-30 13:41:50 |
| SXS:BBH:0005 | True | 1.000460 | 2.498115e-01 | 6.905084e-11 | 2.745050e-09 | 0.000236 | 0.000236 | 390.000000 | -3.646774e-12 | 1.616084e-11 | ... | [-9.01162380399282, -0.0070599738280454, 0.0] | 30.190238 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0] | [Aligned-Spins, ManyMergers-QuasiCircular, dep... | 2012-08-07 19:02:51 | 2012-08-17 05:20:02 | 2012-08-28 09:11:15 | 2019-03-30 13:41:50 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| SXS:NSNS:0006 | False | 1.199984 | 0.000000e+00 | 0.000000e+00 | 0.000000e+00 | 0.000160 | 0.000160 | 363.175161 | 0.000000e+00 | 0.000000e+00 | ... | [20.1881, 0.000300831, 0.0] | NaN | NaN | NaN | NaN | [] | 2022-06-09 20:57:46 | 2022-06-22 14:35:12 | 2022-06-22 14:42:52 | 2023-09-26 23:00:19 |
| SXS:NSNS:0007 | False | 0.831025 | 0.000000e+00 | 0.000000e+00 | 0.000000e+00 | NaN | NaN | 363.158338 | 0.000000e+00 | 0.000000e+00 | ... | [16.3414, -0.000239926, 0.0] | NaN | NaN | NaN | NaN | [] | 2020-04-13 20:55:51 | 2020-11-30 13:36:56 | 2020-11-30 14:14:49 | 2023-10-04 21:50:19 |
| SXS:NSNS:0008 | False | 1.000000 | 0.000000e+00 | 0.000000e+00 | 0.000000e+00 | 0.001317 | 0.001317 | 359.961309 | 0.000000e+00 | 0.000000e+00 | ... | [18.5, -0.000297416, 0.0] | NaN | NaN | NaN | NaN | [] | 2022-06-09 19:14:59 | 2022-06-17 13:10:13 | 2022-06-17 13:17:17 | 2023-09-26 23:00:19 |
| SXS:NSNS:0009 | False | 1.099992 | 0.000000e+00 | 0.000000e+00 | 0.000000e+00 | 0.000920 | 0.000920 | 360.634903 | 0.000000e+00 | 0.000000e+00 | ... | [19.3833, -0.000276636, 0.0] | NaN | NaN | NaN | NaN | [] | 2022-06-09 19:14:59 | 2022-06-17 13:05:03 | 2022-06-17 13:18:31 | 2023-09-26 23:00:19 |
| SXS:NSNS:0010 | False | 1.199984 | 0.000000e+00 | 0.000000e+00 | 0.000000e+00 | 0.000282 | 0.000282 | 301.955378 | 0.000000e+00 | 0.000000e+00 | ... | [20.1875, 0.000277143, 0.0] | NaN | NaN | NaN | NaN | [] | 2022-06-09 19:14:59 | 2022-06-17 13:09:05 | 2022-06-17 13:23:45 | 2023-09-26 23:00:19 |
4170 rows × 87 columns
Most of these columns are described in more detail in the most recent SXS catalog paper — although the dataframe also includes some computed columns, such as the magnitude and individual Cartesian components of any vector.
Note that marimo is a helpful interface (along the lines of Jupyter) with built-in support for pandas objects, including the ability to display interactive tables. Examples are available on the SXS catalog home page at https://data.black-holes.org/.
Slices¶
We can slice the dataframe in a dizzying number of ways. But there are two that are simplest and most reliable. First, and most easily, we can take standard slices, like the first four rows:
In [3]:
Copied!
df[:4]
df[:4]
Out[3]:
| deprecated | reference_mass_ratio | reference_chi_eff | reference_chi1_perp | reference_chi2_perp | reference_eccentricity | reference_eccentricity_bound | reference_time | reference_dimensionless_spin1_x | reference_dimensionless_spin1_y | ... | initial_position2 | number_of_orbits | number_of_orbits_from_start | number_of_orbits_from_reference_time | DOI_versions | keywords | date_link_earliest | date_run_earliest | date_run_latest | date_postprocessing | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SXS:BBH:0001 | True | 1.000000 | 1.215639e-07 | 9.733309e-10 | 1.429939e-09 | 0.000257 | 0.000257 | 320.0 | 7.097521e-10 | -6.660518e-10 | ... | [-8.999999874197535, -2.86646421e-08, 0.0] | 28.119083 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v1.6, v2.0] | [Non-Spinning, ManyMergers-QuasiCircular, depr... | 2012-07-31 17:17:11 | 2012-08-01 01:46:59 | 2012-08-17 20:48:06 | 2019-03-30 13:41:50 |
| SXS:BBH:0002 | True | 1.000000 | 9.399402e-08 | 7.181763e-10 | 1.473155e-09 | 0.000175 | 0.000175 | 310.0 | 6.507563e-10 | -3.037983e-10 | ... | [-9.49999971865619, 2.0830027e-07, 0.0] | 32.421068 | NaN | NaN | [, v1.1, v1.2, v1.4, v1.5, v1.6, v1.7, v2.0] | [Non-Spinning, ManyMergers-QuasiCircular, depr... | 2012-07-31 17:17:11 | 2012-08-01 02:24:21 | 2012-08-26 06:41:57 | 2019-03-30 13:41:50 |
| SXS:BBH:0003 | True | 1.000440 | 2.707179e-04 | 4.993912e-01 | 7.428880e-08 | 0.000287 | 0.000287 | 210.0 | 4.965977e-01 | 5.274908e-02 | ... | [-9.507365032939443, -0.0023408782825215, 0.0] | 32.337574 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0] | [Precessing-Spins, ManyMergers-QuasiCircular, ... | 2012-08-07 19:02:51 | 2012-08-18 20:24:58 | 2012-08-22 10:48:06 | 2019-03-30 13:41:50 |
| SXS:BBH:0004 | True | 1.000459 | -2.498083e-01 | 5.018106e-11 | 1.402755e-09 | 0.000380 | 0.000380 | 640.0 | -2.424428e-12 | -1.449169e-11 | ... | [-9.50286409695725, 0.0031879777731917, 0.0] | 30.188749 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0] | [Aligned-Spins, ManyMergers-QuasiCircular, dep... | 2012-08-07 19:02:51 | 2012-08-18 20:24:58 | 2012-08-25 10:38:13 | 2019-03-30 13:41:50 |
4 rows × 87 columns
Or we can select columns to extract:
In [4]:
Copied!
df[["object_types", "initial_adot"]]
df[["object_types", "initial_adot"]]
Out[4]:
| object_types | initial_adot | |
|---|---|---|
| SXS:BBH:0001 | BHBH | -0.000025 |
| SXS:BBH:0002 | BHBH | -0.000021 |
| SXS:BBH:0003 | BHBH | -0.000019 |
| SXS:BBH:0004 | BHBH | -0.000028 |
| SXS:BBH:0005 | BHBH | -0.000016 |
| ... | ... | ... |
| SXS:NSNS:0006 | NSNS | -0.000027 |
| SXS:NSNS:0007 | NSNS | -0.000030 |
| SXS:NSNS:0008 | NSNS | -0.000031 |
| SXS:NSNS:0009 | NSNS | -0.000032 |
| SXS:NSNS:0010 | NSNS | -0.000028 |
4170 rows × 2 columns
To combine them, we just do them in sequence:
In [5]:
Copied!
df[:4][["object_types", "initial_adot"]]
df[:4][["object_types", "initial_adot"]]
Out[5]:
| object_types | initial_adot | |
|---|---|---|
| SXS:BBH:0001 | BHBH | -0.000025 |
| SXS:BBH:0002 | BHBH | -0.000021 |
| SXS:BBH:0003 | BHBH | -0.000019 |
| SXS:BBH:0004 | BHBH | -0.000028 |
Tests¶
The concept of tests is fairly simple. For example, we can test whether or not the object_types field is equal to BHNS:
In [6]:
Copied!
df["object_types"] == "BHNS"
df["object_types"] == "BHNS"
Out[6]:
SXS:BBH:0001 False
SXS:BBH:0002 False
SXS:BBH:0003 False
SXS:BBH:0004 False
SXS:BBH:0005 False
...
SXS:NSNS:0006 False
SXS:NSNS:0007 False
SXS:NSNS:0008 False
SXS:NSNS:0009 False
SXS:NSNS:0010 False
Name: object_types, Length: 4170, dtype: bool
We get a pandas Series object, where most of the results say False, but the last few say True — because they are the ones for which the object_types field is BHNS. Now, we can use this Series just like we would in numpy to extract the items where this test gives us True:
In [7]:
Copied!
df[df["object_types"] == "BHNS"]
df[df["object_types"] == "BHNS"]
Out[7]:
| deprecated | reference_mass_ratio | reference_chi_eff | reference_chi1_perp | reference_chi2_perp | reference_eccentricity | reference_eccentricity_bound | reference_time | reference_dimensionless_spin1_x | reference_dimensionless_spin1_y | ... | initial_position2 | number_of_orbits | number_of_orbits_from_start | number_of_orbits_from_reference_time | DOI_versions | keywords | date_link_earliest | date_run_earliest | date_run_latest | date_postprocessing | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SXS:BHNS:0001 | False | 6.000024 | 0.000007 | 1.374112e-08 | 0.0 | NaN | 0.001300 | 300.000000 | -1.110994e-08 | -8.083254e-09 | ... | [-8.999999874197535, -2.86646421e-08, 0.0] | 12.647523 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v2.0] | [Non-Spinning, Polytrope] | 2013-03-06 21:52:11 | 2013-03-07 08:12:43 | 2013-04-29 11:11:15 | 2019-04-15 02:48:30 |
| SXS:BHNS:0002 | False | 1.999998 | 0.000002 | 1.662024e-09 | 0.0 | NaN | 0.000500 | 400.000000 | 7.870209e-10 | -1.463872e-09 | ... | [-37.330292064606674, -0.00015284676006403007,... | 13.300000 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v2.0] | [Non-Spinning, Polytrope] | NaT | NaT | NaT | NaT |
| SXS:BHNS:0003 | False | 2.999984 | NaN | NaN | NaN | NaN | 0.000500 | 700.000000 | -6.571999e-08 | -1.563577e-08 | ... | [-38.4325, 0.0, 0.0] | 13.300000 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v2.0] | [Non-Spinning, Polytrope] | NaT | NaT | NaT | NaT |
| SXS:BHNS:0004 | False | 1.000031 | NaN | NaN | NaN | NaN | 0.000060 | 500.000000 | 2.852307e-09 | 9.857248e-10 | ... | [-19.6, 0.0, 0.0] | 12.300000 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0] | [NonSpinning, Polytrope] | NaT | NaT | NaT | NaT |
| SXS:BHNS:0005 | False | 1.000014 | NaN | NaN | NaN | 0.000500 | 0.000500 | 481.500000 | -9.091780e-08 | 2.217490e-09 | ... | [-19.6, 0.0, 0.0] | 10.800000 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0] | [SpinningNS, Polytrope] | NaT | NaT | NaT | NaT |
| SXS:BHNS:0006 | False | 1.500007 | NaN | NaN | NaN | NaN | 0.000270 | 500.000000 | -7.003152e-09 | 1.067139e-08 | ... | [-31.5, 0.0, 0.0] | 16.600000 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0] | [NonSpinning, Polytrope] | NaT | NaT | NaT | NaT |
| SXS:BHNS:0007 | False | 2.000000 | NaN | NaN | NaN | NaN | 0.000470 | 500.000000 | -5.265540e-08 | 2.372880e-08 | ... | [-37.3333, 0.0, 0.0] | 12.300000 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0] | [SpinningNS, Polytrope] | NaT | NaT | NaT | NaT |
| SXS:BHNS:0008 | False | 3.001745 | 0.674294 | 1.724298e-07 | 0.0 | 0.000501 | 0.000501 | 684.569779 | -1.265404e-09 | -7.913752e-09 | ... | [-47.25, 0.0, 0.0] | NaN | NaN | NaN | [, v2.0] | [] | NaT | NaT | NaT | NaT |
| SXS:BHNS:0009 | False | 4.002338 | 0.719236 | 8.007617e-08 | 0.0 | 0.001721 | 0.001721 | 852.584912 | 4.429089e-09 | 4.136482e-09 | ... | [-61.6, 0.0, 0.0] | NaN | NaN | NaN | [, v2.0] | [] | NaT | NaT | NaT | NaT |
| SXS:BHNS:0010 | False | 3.000562 | 0.395555 | 5.321927e-01 | 0.0 | 0.003001 | 0.003001 | 684.577207 | 5.267177e-01 | 4.443504e-02 | ... | [-53.775, 0.0, 0.0] | NaN | NaN | NaN | [, v2.0] | [] | NaT | NaT | NaT | NaT |
10 rows × 87 columns
(Here, we're just looking at the data, so we don't bother with the fancy grid we used above.)
Next, we might want to combine tests. This is done by putting each test inside parentheses, and combinging results with &:
In [8]:
Copied!
df[(df["object_types"] == "BHNS") & (df["initial_separation"] < 52)]
df[(df["object_types"] == "BHNS") & (df["initial_separation"] < 52)]
Out[8]:
| deprecated | reference_mass_ratio | reference_chi_eff | reference_chi1_perp | reference_chi2_perp | reference_eccentricity | reference_eccentricity_bound | reference_time | reference_dimensionless_spin1_x | reference_dimensionless_spin1_y | ... | initial_position2 | number_of_orbits | number_of_orbits_from_start | number_of_orbits_from_reference_time | DOI_versions | keywords | date_link_earliest | date_run_earliest | date_run_latest | date_postprocessing | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SXS:BHNS:0003 | False | 2.999984 | NaN | NaN | NaN | NaN | 0.00050 | 700.0 | -6.571999e-08 | -1.563577e-08 | ... | [-38.4325, 0.0, 0.0] | 13.3 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v2.0] | [Non-Spinning, Polytrope] | NaT | NaT | NaT | NaT |
| SXS:BHNS:0004 | False | 1.000031 | NaN | NaN | NaN | NaN | 0.00006 | 500.0 | 2.852307e-09 | 9.857248e-10 | ... | [-19.6, 0.0, 0.0] | 12.3 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0] | [NonSpinning, Polytrope] | NaT | NaT | NaT | NaT |
| SXS:BHNS:0005 | False | 1.000014 | NaN | NaN | NaN | 0.0005 | 0.00050 | 481.5 | -9.091780e-08 | 2.217490e-09 | ... | [-19.6, 0.0, 0.0] | 10.8 | NaN | NaN | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0] | [SpinningNS, Polytrope] | NaT | NaT | NaT | NaT |
3 rows × 87 columns
Here, the combined test is only True if both tests to return True — the & operator is the boolean AND. We also have OR with | and XOR with ^, as well as negation with ~ — though this can usually be achieved by changing the test.
Built-in filters¶
The df object has various built-in filters that can be used to
extract data. This is because df is actually a
SimulationsDataFrame, which is a subclass of pandas.DataFrame, so
it has extra attributes. We can list them all here:
In [9]:
Copied!
import pandas as pd
set(dir(type(df))) - set(dir(pd.DataFrame))
import pandas as pd
set(dir(type(df))) - set(dir(pd.DataFrame))
Out[9]:
{'BBH',
'BHBH',
'BHNS',
'BNS',
'IMR',
'NSBH',
'NSNS',
'deprecated',
'eccentric',
'hyperbolic',
'noneccentric',
'nonprecessing',
'precessing',
'undeprecated'}
These names are all fairly self-explanatory, and are individually
documented with more details
here.
Each one returns another SimulationsDataFrame with just the selected
simulations; as such they can be combined. For example, to get only
un-deprecated eccentric BBH systems, we can call
df.undeprecated.BBH.eccentric
Sorting¶
As with the fancy graphical table above, we can perform a standard sort with respect to any key:
In [10]:
Copied!
df.BBH.sort_values("initial_separation")
df.BBH.sort_values("initial_separation")
Out[10]:
| deprecated | reference_mass_ratio | reference_chi_eff | reference_chi1_perp | reference_chi2_perp | reference_eccentricity | reference_eccentricity_bound | reference_time | reference_dimensionless_spin1_x | reference_dimensionless_spin1_y | ... | initial_position2 | number_of_orbits | number_of_orbits_from_start | number_of_orbits_from_reference_time | DOI_versions | keywords | date_link_earliest | date_run_earliest | date_run_latest | date_postprocessing | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SXS:BBH:1145 | False | 1.249997 | 2.144264e-06 | 3.898542e-10 | 3.458803e-11 | NaN | NaN | 13.0 | -3.829624e-10 | -7.298039e-11 | ... | [-4.871504224452922, 0.0126460374892269, 0.0] | 3.495635e+00 | 3.495635e+00 | 3.433299e+00 | [, v1.2, v1.3, v1.4, v1.5, v2.0, v3.0] | [] | 2015-09-23 15:06:10 | 2015-09-23 18:38:59 | 2015-09-26 22:34:35 | 2025-01-28 01:00:31.247358 |
| SXS:BBH:3722 | False | 3.817251 | 3.335513e-01 | 8.981276e-01 | 7.176135e-01 | 0.152158 | 0.152158 | 342.5 | 8.546314e-01 | 3.483516e-02 | ... | [-7.451739902660199, 0.035700869734113, 0.0334... | 9.975052e+00 | 9.975052e+00 | 8.350220e+00 | [, v2.0, v3.0] | [] | 2022-01-15 21:49:17 | 2022-03-04 17:23:06 | 2022-07-16 03:33:22 | 2025-01-28 18:15:26.418029 |
| SXS:BBH:3704 | False | 1.666655 | 3.074859e-01 | 9.911856e-03 | 3.157166e-04 | NaN | NaN | 240.0 | 8.447812e-03 | 4.831329e-03 | ... | [-5.8973114013009935, 0.0178116687187507, 0.00... | 6.396565e+00 | 6.396565e+00 | 5.182133e+00 | [, v2.0, v3.0] | [] | 2020-06-18 05:27:31 | 2020-07-19 12:46:31 | 2020-08-06 17:18:07 | 2025-01-28 16:40:21.204794 |
| SXS:BBH:3726 | False | 1.228916 | 2.937017e-01 | 8.563209e-01 | 3.719536e-01 | 0.270289 | 0.270289 | 364.5 | -7.421380e-01 | -4.340325e-01 | ... | [-5.1821982922704874, 0.0101374549239111, -0.0... | 8.439590e+00 | 8.439590e+00 | 6.888812e+00 | [, v2.0, v3.0] | [] | 2021-07-21 19:36:21 | 2022-01-11 18:34:14 | 2022-02-08 03:23:22 | 2025-01-28 18:14:53.938003 |
| SXS:BBH:3720 | False | 1.240874 | 2.682820e-01 | 7.159846e-01 | 7.572758e-01 | NaN | NaN | 301.5 | -5.061743e-01 | -4.474765e-01 | ... | [-5.252781561663051, 0.0089879732395598, -0.03... | 6.843151e+00 | 6.843151e+00 | 5.407691e+00 | [, v2.0, v3.0] | [] | 2022-01-15 21:49:17 | 2022-02-18 11:30:03 | 2022-03-04 17:35:56 | 2025-01-28 18:14:46.492079 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| SXS:BBH:4000 | False | 1.000000 | 2.353748e-10 | 3.186150e-10 | 3.150481e-10 | NaN | NaN | 670.0 | 3.366684e-11 | 3.168313e-10 | ... | [-125.00007703959201, 1.57972238326e-05, 0.000... | 3.838468e+00 | 3.838468e+00 | 3.828192e+00 | [, v3.0] | [] | 2024-07-25 08:17:07 | 2024-07-25 11:13:06 | 2024-08-04 06:32:14 | 2025-01-28 16:31:02.163013 |
| SXS:BBH:3990 | True | 1.000000 | -1.349331e-14 | 2.340154e-10 | 2.311762e-10 | NaN | NaN | 310.0 | -2.318374e-13 | 2.340779e-10 | ... | [-199.999521558393, -0.0001099573632147, -8.63... | 9.753268e-09 | 9.753268e-09 | 9.723255e-09 | [, v3.0] | [deprecated] | 2024-09-15 22:58:33 | 2024-09-20 09:39:59 | 2024-09-26 09:12:45 | 2025-04-08 20:13:56.568156 |
| SXS:BBH:3996 | False | 1.000000 | 7.517316e-14 | 2.351166e-10 | 2.310230e-10 | NaN | NaN | 310.0 | -2.378321e-13 | 2.351841e-10 | ... | [-199.999521558393, -0.0001099573632147, -8.63... | 9.706578e-09 | 9.706578e-09 | 9.676564e-09 | [, v3.0] | [] | 2024-10-16 15:35:58 | 2024-10-18 16:13:23 | 2024-11-13 06:55:19 | 2025-01-28 18:19:35.519535 |
| SXS:BBH:3987 | True | 1.000000 | 4.176085e-14 | 2.341179e-10 | 2.308822e-10 | NaN | NaN | 310.0 | -2.209407e-13 | 2.341826e-10 | ... | [-199.999521558393, -0.0001099573632147, -8.63... | 9.938248e-09 | 9.938248e-09 | 9.908231e-09 | [, v3.0] | [deprecated] | 2024-09-09 16:48:50 | 2024-09-10 10:46:35 | 2024-09-21 01:05:14 | 2025-04-08 20:10:11.081591 |
| SXS:BBH:3993 | True | 1.000000 | -4.327871e-14 | 2.336804e-10 | 2.313359e-10 | NaN | NaN | 310.0 | -2.206551e-13 | 2.337447e-10 | ... | [-199.999521558393, -0.0001099573632147, -8.63... | 9.766881e-09 | 9.766881e-09 | 9.736864e-09 | [, v3.0] | [deprecated] | 2024-09-09 16:48:50 | 2024-09-10 10:56:00 | 2024-09-20 02:56:53 | 2025-04-08 19:57:43.249388 |
4150 rows × 87 columns
But unlike the fancy graphical table above, we can use a function that serves as the sort key. (This sort of key function is also available in the standard python library's sorted function.) Here, we'll sort by the absolute value of the difference between initial_separation and 20.0.
In [11]:
Copied!
sorting_field = "initial_separation"
desired_value = 20.0
df.BBH.sort_values(sorting_field, key=lambda s: abs(s-desired_value))
sorting_field = "initial_separation"
desired_value = 20.0
df.BBH.sort_values(sorting_field, key=lambda s: abs(s-desired_value))
Out[11]:
| deprecated | reference_mass_ratio | reference_chi_eff | reference_chi1_perp | reference_chi2_perp | reference_eccentricity | reference_eccentricity_bound | reference_time | reference_dimensionless_spin1_x | reference_dimensionless_spin1_y | ... | initial_position2 | number_of_orbits | number_of_orbits_from_start | number_of_orbits_from_reference_time | DOI_versions | keywords | date_link_earliest | date_run_earliest | date_run_latest | date_postprocessing | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SXS:BBH:0204 | False | 7.001168 | 3.498908e-01 | 6.217940e-08 | 2.747613e-08 | 0.000026 | 0.000026 | 296.0 | 2.292678e-08 | 3.490818e-08 | ... | [-17.515569995515307, 0.0093443633956847, 0.0] | 8.836930e+01 | 8.836930e+01 | 8.787693e+01 | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0, v3.0] | [] | 2013-08-23 22:48:09 | 2013-09-02 15:05:40 | 2014-04-01 19:53:23 | 2025-04-16 20:03:07.176331 |
| SXS:BBH:1130 | False | 1.000004 | -5.104350e-04 | 5.999922e-01 | 1.560264e-08 | 0.000252 | 0.000252 | 160.0 | -3.991107e-02 | 5.986615e-01 | ... | [-9.986900818249802, -0.0079228234220203, 0.0] | 3.677521e+01 | 3.677521e+01 | 3.650749e+01 | [, v1.2, v1.3, v1.4, v1.5, v2.0, v3.0] | [] | 2014-07-16 18:32:05 | 2014-07-21 08:37:28 | 2014-09-15 16:10:45 | 2025-01-28 03:39:39.749483 |
| SXS:BBH:1131 | False | 1.000017 | 8.786219e-04 | 5.999634e-01 | 1.941851e-08 | 0.000087 | 0.000087 | 470.0 | 5.841738e-01 | 1.367446e-01 | ... | [-10.101294991597507, -0.010416650565165, 0.0] | 3.679601e+01 | 3.679601e+01 | 3.600615e+01 | [, v1.2, v1.3, v1.4, v1.5, v2.0, v3.0] | [] | 2014-06-30 17:56:22 | 2014-07-17 16:56:33 | 2014-10-02 20:35:34 | 2025-01-28 03:37:18.462033 |
| SXS:BBH:0623 | False | 1.111081 | 5.742299e-03 | 8.917771e-01 | 8.915548e-01 | 0.000067 | 0.000067 | 230.0 | 8.864194e-01 | 1.012121e-01 | ... | [-10.537514026848347, -0.0398791824195513, 0.0] | 3.747046e+01 | 3.747046e+01 | 3.708570e+01 | [, v1.2, v1.3, v1.4, v1.5, v2.0, v3.0] | [] | 2015-01-09 00:07:43 | 2015-01-09 02:04:36 | 2016-06-07 08:58:47 | 2025-01-28 03:44:07.649317 |
| SXS:BBH:3661 | True | 1.400062 | 3.288885e-01 | 2.950220e-08 | 5.486936e-08 | 0.000065 | 0.000065 | 1494.5 | 6.919824e-08 | -3.278550e-08 | ... | [-11.783346845755291, -0.0402772562166922, 0.0] | 4.100697e+01 | 4.100697e+01 | 3.847501e+01 | [, v2.0, v3.0] | [deprecated] | 2015-04-06 20:49:13 | 2015-04-28 15:57:48 | 2015-07-26 06:28:36 | 2025-01-28 18:25:21.177337 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| SXS:BBH:4000 | False | 1.000000 | 2.353748e-10 | 3.186150e-10 | 3.150481e-10 | NaN | NaN | 670.0 | 3.366684e-11 | 3.168313e-10 | ... | [-125.00007703959201, 1.57972238326e-05, 0.000... | 3.838468e+00 | 3.838468e+00 | 3.828192e+00 | [, v3.0] | [] | 2024-07-25 08:17:07 | 2024-07-25 11:13:06 | 2024-08-04 06:32:14 | 2025-01-28 16:31:02.163013 |
| SXS:BBH:3990 | True | 1.000000 | -1.349331e-14 | 2.340154e-10 | 2.311762e-10 | NaN | NaN | 310.0 | -2.318374e-13 | 2.340779e-10 | ... | [-199.999521558393, -0.0001099573632147, -8.63... | 9.753268e-09 | 9.753268e-09 | 9.723255e-09 | [, v3.0] | [deprecated] | 2024-09-15 22:58:33 | 2024-09-20 09:39:59 | 2024-09-26 09:12:45 | 2025-04-08 20:13:56.568156 |
| SXS:BBH:3996 | False | 1.000000 | 7.517316e-14 | 2.351166e-10 | 2.310230e-10 | NaN | NaN | 310.0 | -2.378321e-13 | 2.351841e-10 | ... | [-199.999521558393, -0.0001099573632147, -8.63... | 9.706578e-09 | 9.706578e-09 | 9.676564e-09 | [, v3.0] | [] | 2024-10-16 15:35:58 | 2024-10-18 16:13:23 | 2024-11-13 06:55:19 | 2025-01-28 18:19:35.519535 |
| SXS:BBH:3993 | True | 1.000000 | -4.327871e-14 | 2.336804e-10 | 2.313359e-10 | NaN | NaN | 310.0 | -2.206551e-13 | 2.337447e-10 | ... | [-199.999521558393, -0.0001099573632147, -8.63... | 9.766881e-09 | 9.766881e-09 | 9.736864e-09 | [, v3.0] | [deprecated] | 2024-09-09 16:48:50 | 2024-09-10 10:56:00 | 2024-09-20 02:56:53 | 2025-04-08 19:57:43.249388 |
| SXS:BBH:3987 | True | 1.000000 | 4.176085e-14 | 2.341179e-10 | 2.308822e-10 | NaN | NaN | 310.0 | -2.209407e-13 | 2.341826e-10 | ... | [-199.999521558393, -0.0001099573632147, -8.63... | 9.938248e-09 | 9.938248e-09 | 9.908231e-09 | [, v3.0] | [deprecated] | 2024-09-09 16:48:50 | 2024-09-10 10:46:35 | 2024-09-21 01:05:14 | 2025-04-08 20:10:11.081591 |
4150 rows × 87 columns
So, if we want the 8 systems with initial separations closest to 20, we can just take them:
In [12]:
Copied!
sorting_field = "initial_separation"
desired_value = 20.0
N = 8
df.BBH.sort_values(sorting_field, key=lambda s: abs(s-desired_value))[:N]
sorting_field = "initial_separation"
desired_value = 20.0
N = 8
df.BBH.sort_values(sorting_field, key=lambda s: abs(s-desired_value))[:N]
Out[12]:
| deprecated | reference_mass_ratio | reference_chi_eff | reference_chi1_perp | reference_chi2_perp | reference_eccentricity | reference_eccentricity_bound | reference_time | reference_dimensionless_spin1_x | reference_dimensionless_spin1_y | ... | initial_position2 | number_of_orbits | number_of_orbits_from_start | number_of_orbits_from_reference_time | DOI_versions | keywords | date_link_earliest | date_run_earliest | date_run_latest | date_postprocessing | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SXS:BBH:0204 | False | 7.001168 | 3.498908e-01 | 6.217940e-08 | 2.747613e-08 | 0.000026 | 0.000026 | 296.0 | 2.292678e-08 | 3.490818e-08 | ... | [-17.515569995515307, 0.0093443633956847, 0.0] | 88.369299 | 88.369299 | 87.876931 | [, v1.1, v1.2, v1.3, v1.4, v1.5, v2.0, v3.0] | [] | 2013-08-23 22:48:09 | 2013-09-02 15:05:40 | 2014-04-01 19:53:23 | 2025-04-16 20:03:07.176331 |
| SXS:BBH:1130 | False | 1.000004 | -5.104350e-04 | 5.999922e-01 | 1.560264e-08 | 0.000252 | 0.000252 | 160.0 | -3.991107e-02 | 5.986615e-01 | ... | [-9.986900818249802, -0.0079228234220203, 0.0] | 36.775214 | 36.775214 | 36.507493 | [, v1.2, v1.3, v1.4, v1.5, v2.0, v3.0] | [] | 2014-07-16 18:32:05 | 2014-07-21 08:37:28 | 2014-09-15 16:10:45 | 2025-01-28 03:39:39.749483 |
| SXS:BBH:1131 | False | 1.000017 | 8.786219e-04 | 5.999634e-01 | 1.941851e-08 | 0.000087 | 0.000087 | 470.0 | 5.841738e-01 | 1.367446e-01 | ... | [-10.101294991597507, -0.010416650565165, 0.0] | 36.796011 | 36.796011 | 36.006148 | [, v1.2, v1.3, v1.4, v1.5, v2.0, v3.0] | [] | 2014-06-30 17:56:22 | 2014-07-17 16:56:33 | 2014-10-02 20:35:34 | 2025-01-28 03:37:18.462033 |
| SXS:BBH:0623 | False | 1.111081 | 5.742299e-03 | 8.917771e-01 | 8.915548e-01 | 0.000067 | 0.000067 | 230.0 | 8.864194e-01 | 1.012121e-01 | ... | [-10.537514026848347, -0.0398791824195513, 0.0] | 37.470458 | 37.470458 | 37.085702 | [, v1.2, v1.3, v1.4, v1.5, v2.0, v3.0] | [] | 2015-01-09 00:07:43 | 2015-01-09 02:04:36 | 2016-06-07 08:58:47 | 2025-01-28 03:44:07.649317 |
| SXS:BBH:3661 | True | 1.400062 | 3.288885e-01 | 2.950220e-08 | 5.486936e-08 | 0.000065 | 0.000065 | 1494.5 | 6.919824e-08 | -3.278550e-08 | ... | [-11.783346845755291, -0.0402772562166922, 0.0] | 41.006968 | 41.006968 | 38.475014 | [, v2.0, v3.0] | [deprecated] | 2015-04-06 20:49:13 | 2015-04-28 15:57:48 | 2015-07-26 06:28:36 | 2025-01-28 18:25:21.177337 |
| SXS:BBH:2299 | True | 1.000000 | -2.063040e-06 | 1.002663e-09 | 9.446935e-10 | 0.280834 | 0.280834 | 160.0 | 2.961036e-10 | -9.579443e-10 | ... | [-9.998901474327717, 1.198447144e-07, 3.879346... | 28.242072 | NaN | NaN | [, v2.0] | deprecated | 2019-07-23 23:57:41 | 2019-08-16 02:51:53 | 2019-08-22 10:37:56 | 2019-08-23 18:56:35.000000 |
| SXS:BBH:2603 | False | 1.000000 | -2.031117e-06 | 2.287214e-09 | 3.221655e-09 | 0.283301 | 0.283301 | 410.0 | 6.431297e-11 | -2.286310e-09 | ... | [-9.998901119028679, 2.38530902e-07, -3.957805... | 28.239080 | 28.239080 | 27.674731 | [, v2.0, v3.0] | [] | 2021-04-15 03:34:04 | 2021-04-15 18:35:46 | 2021-05-12 19:24:55 | 2025-01-28 15:09:48.594719 |
| SXS:BBH:2560 | False | 8.000056 | -1.033673e-07 | 8.031537e-08 | 3.305479e-09 | 0.186632 | 0.186632 | 230.0 | 2.038238e-08 | 7.768602e-08 | ... | [-17.789197381375974, 0.0067012546281216, -1.0... | 53.085495 | 53.085495 | 52.645781 | [, v3.0] | [] | 2020-10-28 09:07:01 | 2020-11-09 18:47:46 | 2021-02-25 00:42:47 | 2025-01-29 21:27:33.691318 |
8 rows × 87 columns
Plotting¶
Pandas also makes it easy to plot the various quantities. For example, we can make a scatter plot of mass ratio versus $\chi_{\mathrm{eff}}$:
In [13]:
Copied!
df.BBH.plot("reference_mass_ratio", "reference_chi_eff", kind="scatter")
# pandas adds the column labels as axis labels, but we can make them look nicer
import matplotlib.pyplot as plt
plt.xlabel(r"Mass ratio")
plt.ylabel(r"$\chi_\mathrm{eff}$");
df.BBH.plot("reference_mass_ratio", "reference_chi_eff", kind="scatter")
# pandas adds the column labels as axis labels, but we can make them look nicer
import matplotlib.pyplot as plt
plt.xlabel(r"Mass ratio")
plt.ylabel(r"$\chi_\mathrm{eff}$");
Or we can make histograms of the data:
In [14]:
Copied!
df.BBH["initial_ADM_linear_momentum_mag"].plot.hist(log=True)
plt.xlabel(r"Magnitude of ADM linear momentum");
df.BBH["initial_ADM_linear_momentum_mag"].plot.hist(log=True)
plt.xlabel(r"Magnitude of ADM linear momentum");
We can even make corner plots:
In [15]:
Copied!
import seaborn as sns
pp = sns.pairplot(
df.BBH[["reference_chi_eff", "reference_chi1_perp", "reference_chi2_perp", "reference_mass_ratio"]].dropna(),
corner=True,
)
pp.y_vars = [r"$\chi_{\mathrm{eff}}$", r"$\chi_{\perp,1}$", r"$\chi_{\perp,2}$", r"$q$"]
pp.x_vars = pp.y_vars
pp._add_axis_labels()
import seaborn as sns
pp = sns.pairplot(
df.BBH[["reference_chi_eff", "reference_chi1_perp", "reference_chi2_perp", "reference_mass_ratio"]].dropna(),
corner=True,
)
pp.y_vars = [r"$\chi_{\mathrm{eff}}$", r"$\chi_{\perp,1}$", r"$\chi_{\perp,2}$", r"$q$"]
pp.x_vars = pp.y_vars
pp._add_axis_labels()
The simulations object, especially when agumented with the dataframe, provides powerful methods for selecting the particular simulations we are interested in. Once we have done so, we need to load and interact with the simulations.
Continue with the introduction to the Simulation objects.