smilegeng commited on
Commit
82d7d4d
·
verified ·
1 Parent(s): f77179e

Rewrite dataset card in English; add tsfile tag

Browse files
Files changed (1) hide show
  1. README.md +52 -40
README.md CHANGED
@@ -1,40 +1,52 @@
1
- # REG-Forecasting_v2 (TsFile 格式)
2
-
3
- > 原数据集:https://huggingface.co/datasets/wachawich/REG-Forecasting_v2
4
- > 许可证:**原数据集未声明许可证**,使用请以原数据集为准。
5
-
6
- 本仓库是 `wachawich/REG-Forecasting_v2` 转换为 Apache TsFile 格式的版本,用于时序数据库 / 时序预测场景。
7
-
8
- ## 数据内容
9
-
10
- 单一地区的可再生能源发电预测数据,时间范围 **2020-01-01 ~ 2025-11-19**,逐小时采样。每个小时有两种发电类型(Solar、Wind)的发电量,以及共享的气象与天文特征。
11
-
12
- ## TsFile 存储模型
13
-
14
- - **表名**:`reg_forecasting_v2`
15
- - **TAG(设备维度)**:`type_name`(`Solar` / `Wind`)—— 每种发电类型是一条独立时间序列
16
- - **Time**:由原始 `date`(YYYY-MM-DD)+ `time`(一天内的小时槽 0~23)**拼接**成真实时间戳,INT64 毫秒
17
- - **FIELD**:`value`(发电量)+ 42 个气象/天文/日期分量特征列(辐射、云量、温度、湿度、气压、风速风向、降水、日出日落、周期正余弦编码等)
18
- - 行数:**103,079**(去重后)
19
-
20
- 类型映射:float64 → DOUBLE,整数 → INT64,字符串 → STRING。
21
-
22
- ## 与原数据集的差异(转换说明)
23
-
24
- 转换过程中以下列被丢弃或变动,**原因如下**:
25
-
26
- | 列 | 处理 | 原因 |
27
- |---|---|---|
28
- | `date` | 丢弃 | 与 `time` 拼接进 `Time` 列,信息无损保留 |
29
- | `time` | 丢弃 | 一天内的小时槽,已拼接进 `Time` 列 |
30
- | `fueltype` | 丢弃 | 与 TAG 列 `type-name` 完全等价(1=Solar, 2=Wind),避免冗余 |
31
- | `type-name` | 重命名为 `type_name` | 原名含连字符,不是合法 TsFile 标识符;作为 TAG 列 |
32
- | 46 个重复行 | 去重 | 夏令时切换日(如 03-31 / 06-29 的 time=0)产生的完全相同重复记录,TsFile 同一设备同一时间戳只能有一个点 |
33
-
34
- 其余 42 个气象/天文/日期分量特征列**全部保留**为 FIELD。
35
-
36
- ## 读回校验
37
-
38
- - 行数:staged parquet 103,079 == tsfile 读回 103,079 ✓
39
- - tag:`type_name` category=TAG,其余 43 列 category=FIELD ✓
40
- - 时间戳拼接:首行 `1577836800000` = 2020-01-01 00:00:00 UTC,次行 +1 小时 ✓
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - tsfile
4
+ - time-series
5
+ - tabular
6
+ - energy
7
+ - forecasting
8
+ - weather
9
+ pretty_name: REG-Forecasting_v2
10
+ size_categories:
11
+ - 100K<n<1M
12
+ ---
13
+
14
+ # REG-Forecasting_v2 (TsFile)
15
+
16
+ Apache TsFile version of
17
+ [`wachawich/REG-Forecasting_v2`](https://huggingface.co/datasets/wachawich/REG-Forecasting_v2).
18
+
19
+ ## Overview
20
+
21
+ Renewable Energy Generation (REG) forecasting data at **hourly** granularity, spanning
22
+ **2020-01-01 to 2025-11-19**. For each hour it records the generation (`value`) of two
23
+ generation types together with the shared meteorological and solar-geometry features.
24
+ Compared with the v1 dataset, this version adds extra columns such as `solar_count`,
25
+ `wind_turbine_count`, and cyclic time encodings (`sin_time`, `cos_time`,
26
+ `day_of_month_sin/cos`, `month_of_year_sin/cos`).
27
+
28
+ - **Two generation types:** `type_name` = `Solar` / `Wind`.
29
+ - **Forecast target:** `value` — generation in that hour.
30
+ - **Features:** radiation, cloud cover, temperature, humidity, pressure, wind
31
+ speed/direction, precipitation, sunrise/sunset times, and cyclic calendar encodings.
32
+
33
+ ## Schema (TsFile structure)
34
+
35
+ - **Time** (INT64, milliseconds) — the hourly timestamp.
36
+ - **type_name** (TAG) — the device dimension, `Solar` / `Wind`. Query one type with
37
+ `WHERE type_name='Solar'`.
38
+ - All remaining columns are FIELDs: `value` plus the meteorological and calendar features.
39
+
40
+ Type mapping: floating-point → DOUBLE, integers → INT64, strings → STRING. A small number
41
+ of byte-for-byte duplicate rows (from daylight-saving transitions) were removed so that a
42
+ single device never carries two points at the same timestamp.
43
+
44
+ ## Usage
45
+
46
+ Read `data.tsfile` with the Apache TsFile Java or Python SDK.
47
+
48
+ ## Source & license
49
+
50
+ - Original dataset: https://huggingface.co/datasets/wachawich/REG-Forecasting_v2
51
+ - Author: wachawich
52
+ - License: not declared by the original dataset card; please defer to the original dataset.