juliensimon/kepler-transit-timing

---
license: cc-by-4.0
pretty_name: "Kepler Transit Timing Catalog"
language:
- en
description: "Holczer et al. (2016) Kepler transit timing catalog — 295,187 individual transit times for 2,599 Kepler Objects of Interest (KOIs), with O-C residuals, durations, and depths."
task_categories:
- tabular-regression
tags:
- space
- exoplanets
- kepler
- transit-timing
- ttv
- astronomy
- open-data
- tabular-data
size_categories:
- 100K<n<1M
configs:
- config_name: default
data_files:
- split: train
path: data/kepler_transit_timing.parquet
default: true
---

# Kepler Transit Timing Catalog

*Part of the [Astronomy Datasets](https://huggingface.co/collections/juliensimon/astronomy-datasets-69c24caf2f17e36128946743) collection on Hugging Face.*

Transit timing catalog from Holczer et al. (2016), containing **295,187** individual transit
mid-times for **2,599** Kepler Objects of Interest (KOIs). Each record includes the
observed mid-transit time, observed-minus-computed (O-C) residual, transit duration, and
transit depth with uncertainties.

## Dataset description

Transit timing variations (TTVs) occur when gravitational interactions between planets in a
multi-planet system cause measurable deviations from a strictly periodic transit schedule.
Holczer et al. (2016) performed a uniform analysis of all Kepler long-cadence light curves
to extract individual transit times, producing the most comprehensive Kepler TTV catalog.
The O-C (observed minus computed) residuals reveal planetary interactions, orbital
eccentricities, and the presence of additional non-transiting planets.

## Key columns

| Column | Type | Description |
|--------|------|-------------|
| `koi` | float64 | Kepler Object of Interest number |
| `transit_number` | Int32 | Sequential transit number for this KOI |
| `t_obs_bjd` | float64 | Observed mid-transit time (BJD - 2454833) |
| `t_obs_err` | float64 | Uncertainty on mid-transit time (days) |
| `o_c` | float64 | Observed minus computed residual (days) |
| `o_c_err` | float64 | Uncertainty on O-C residual (days) |
| `duration_hr` | float64 | Transit duration (hours) |
| `duration_err` | float64 | Uncertainty on transit duration (hours) |
| `depth_ppm` | float64 | Transit depth (ppm) |
| `depth_err` | float64 | Uncertainty on transit depth (ppm) |

## Quick stats

- **295,187** individual transit times
- **2,599** unique KOIs
- Median O-C residual: **0.0000** days
- Median transit depth: **nan** ppm
- Median transit duration: **nan** hours

## Usage

```python
from datasets import load_dataset

ds = load_dataset("juliensimon/kepler-transit-timing", split="train")
df = ds.to_pandas()

# TTVs for a specific KOI
koi_137 = df[df["koi"] == 137.01].sort_values("transit_number")
print(f"KOI 137.01: {len(koi_137)} transits")

# Plot O-C diagram
import matplotlib.pyplot as plt
plt.errorbar(koi_137["transit_number"], koi_137["o_c"],
yerr=koi_137["o_c_err"], fmt=".", ms=3)
plt.xlabel("Transit number")
plt.ylabel("O-C (days)")
plt.title("KOI 137.01 Transit Timing Variations")
plt.show()

# KOIs with the strongest TTVs (largest O-C scatter)
ttv_rms = df.groupby("koi")["o_c"].std().sort_values(ascending=False)
print("Top 10 TTV candidates:")
print(ttv_rms.head(10))
```

## Data source

Holczer, T. et al. (2016), "Transit Timing Observations from Kepler. IX. Catalog of
Transit Timing Measurements of the Long-Cadence Data", ApJS, 225, 9. Accessed via
[VizieR](https://vizier.cds.unistra.fr/), CDS Strasbourg (J/ApJS/225/9).

## Pipeline

Source code: [juliensimon/space-datasets](https://github.com/juliensimon/space-datasets)

## Citation

```bibtex
@dataset{kepler_transit_timing,
author = {Simon, Julien},
title = {Kepler Transit Timing Catalog},
year = {2026},
publisher = {Hugging Face},
url = {https://huggingface.co/datasets/juliensimon/kepler-transit-timing},
note = {Based on Holczer et al. (2016) ApJS 225, 9, via VizieR CDS}
}
```

## License

[CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)

许可证：CC-BY-4.0
规范名称：开普勒凌日时间目录
语言：英语
数据集描述：Holczer等人（2016）发布的开普勒凌日时间目录，包含2599个开普勒目标天体（Kepler Objects of Interest，KOIs）的295187条独立凌日时间数据，附带观测减计算（Observed minus Computed，O-C）残差、凌日时长与凌日深度。
任务类别：表格回归
标签：太空、系外行星、开普勒望远镜、凌日时间、凌日时间变化（Transit Timing Variations，TTVs）、天文学、开放数据、表格数据
数据规模：100K<n<1M
配置项：
- 配置名称：default（默认配置）
数据文件：
- 划分：train（训练集）
路径：data/kepler_transit_timing.parquet
默认为当前配置

# 开普勒凌日时间目录

本数据集隶属于Hugging Face平台上的[Astronomy Datasets](https://huggingface.co/collections/juliensimon/astronomy-datasets-69c24caf2f17e36128946743)数据集集合。

本数据集源自Holczer等人（2016）的研究成果，包含**295,187条独立凌日中时刻数据，对应**2,599个开普勒目标天体（KOIs）。每条记录均包含观测凌日中时刻、观测减计算（O-C）残差、凌日时长以及带不确定度的凌日深度。

## 数据集描述

凌日时间变化（Transit Timing Variations，TTVs）指多行星系统内行星间的引力相互作用，会使严格周期性的凌日现象出现可观测的偏差。Holczer等人（2016）对所有开普勒长周期测光曲线进行了统一分析，提取出各独立凌日时间，构建了目前最全面的开普勒TTV目录。O-C（观测减计算）残差可用于揭示行星间的相互作用、轨道偏心率以及未发生凌日的系外行星的存在。

## 关键列

| 列名 | 数据类型 | 描述 |
|--------|------|-------------|
| `koi` | float64 | 开普勒目标天体编号 |
| `transit_number` | Int32 | 该目标天体的凌日序列编号 |
| `t_obs_bjd` | float64 | 观测到的凌日中时刻（BJD-2454833 |
| `t_obs_err` | float64 | 凌日中时刻的不确定度（单位：天） |
| `o_c` | float64 | 观测减计算残差（单位：天） |
| `o_c_err` | float64 | O-C残差的不确定度（单位：天） |
| `duration_hr` | float64 | 凌日时长（单位：小时） |
| `duration_err` | float64 | 凌日时长的不确定度（单位：小时） |
| `depth_ppm` | float64 | 凌日深度（单位：百万分之一，ppm） |
| `depth_err` | float64 | 凌日深度的不确定度（单位：百万分之一，ppm） |

## 快速统计

- **295,187**条独立凌日时间数据
- **2,599**个唯一开普勒目标天体
- O-C残差中位数：**0.0000**天
- 凌日深度中位数：nan ppm
- 凌日时长中位数：nan 小时

## 使用方法

python
from datasets import load_dataset

ds = load_dataset("juliensimon/kepler-transit-timing", split="train")
df = ds.to_pandas()

# 提取特定KOI的凌日时间变化数据
koi_137 = df[df["koi"] == 137.01].sort_values("transit_number")
print(f"KOI 137.01：共{len(koi_137)}次凌日")

# 绘制O-C残差图
import matplotlib.pyplot as plt
plt.errorbar(koi_137["transit_number"], koi_137["o_c"], yerr=koi_137["o_c_err"], fmt=".", ms=3)
plt.xlabel("凌日序列编号")
plt.ylabel("O-C残差（天）")
plt.title("KOI 137.01 凌日时间变化")
plt.show()

# 筛选凌日时间变化效应最强的KOI（O-C残差离散度最大的天体）
ttv_rms = df.groupby("koi")["o_c"].std().sort_values(ascending=False)
print("凌日时间变化候选天体TOP10：")
print(ttv_rms.head(10))

## 数据来源

Holczer, T. 等人（2016），《开普勒凌日观测 第九期：长周期测光数据凌日时间测量目录》，《天体物理期刊增刊系列》（ApJS），225卷，第9页。数据通过斯特拉斯堡天文数据中心（CDS）的[VizieR](https://vizier.cds.unistra.fr/)平台获取（编号：J/ApJS/225/9）。

## 数据管道

源代码：[juliensimon/space-datasets](https://github.com/juliensimon/space-datasets)

## 引用格式

bibtex
@dataset{kepler_transit_timing,
author = {Simon, Julien},
title = {Kepler Transit Timing Catalog},
year = {2026},
publisher = {Hugging Face},
url = {https://huggingface.co/datasets/juliensimon/kepler-transit-timing},
note = {Based on Holczer et al. (2016) ApJS 225, 9, via VizieR CDS}
}

## 许可证

[CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)