Datasets:
Tasks:
Time Series Forecasting
Modalities:
Time-series
Size:
n<1K
Tags:
language_creators:found
forecasting
benchmark
monash-time-series-forecasting-repository
monash-tsf
tsfile
License:
Update dataset card wording
Browse files
README.md
CHANGED
|
@@ -35,7 +35,7 @@ configs:
|
|
| 35 |
|
| 36 |
18 daily time series including hash rate, block size, mining difficulty etc. as well as public opinion in the form of tweets and google searches mentioning the keyword bitcoin as potential influencer of the bitcoin price.
|
| 37 |
|
| 38 |
-
This repository contains the full source `.tsf` series from the Monash Time Series Forecasting Repository converted to [Apache TsFile](https://tsfile.apache.org/) format.
|
| 39 |
|
| 40 |
## Summary
|
| 41 |
|
|
@@ -73,7 +73,6 @@ This repository contains the full source `.tsf` series from the Monash Time Seri
|
|
| 73 |
- Source `.tsf` attributes are stored as TAG columns.
|
| 74 |
- The `target` series values are flattened into timestamped rows and stored as a FLOAT FIELD.
|
| 75 |
- `Time` is synthesized from the source start timestamp and the `.tsf` frequency metadata, with millisecond precision.
|
| 76 |
-
- The conversion preserves the complete source series and does not create train, validation, or test split files.
|
| 77 |
- Large outputs may be sharded by the TsFile conversion tool; all listed shards belong to the same logical table `bitcoin`.
|
| 78 |
|
| 79 |
## Reading Example
|
|
|
|
| 35 |
|
| 36 |
18 daily time series including hash rate, block size, mining difficulty etc. as well as public opinion in the form of tweets and google searches mentioning the keyword bitcoin as potential influencer of the bitcoin price.
|
| 37 |
|
| 38 |
+
This repository contains the full source `.tsf` series from the Monash Time Series Forecasting Repository converted to [Apache TsFile](https://tsfile.apache.org/) format.
|
| 39 |
|
| 40 |
## Summary
|
| 41 |
|
|
|
|
| 73 |
- Source `.tsf` attributes are stored as TAG columns.
|
| 74 |
- The `target` series values are flattened into timestamped rows and stored as a FLOAT FIELD.
|
| 75 |
- `Time` is synthesized from the source start timestamp and the `.tsf` frequency metadata, with millisecond precision.
|
|
|
|
| 76 |
- Large outputs may be sharded by the TsFile conversion tool; all listed shards belong to the same logical table `bitcoin`.
|
| 77 |
|
| 78 |
## Reading Example
|