Datasets:

smilegeng
/

airespucrs_sales

Exception:    SplitsNotFoundError
Message:      The split names could not be parsed from the dataset config.
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.12/site-packages/datasets/inspect.py", line 286, in get_dataset_config_info
                  for split_generator in builder._split_generators(
                                         ^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/tsfile/tsfile.py", line 271, in _split_generators
                  scan = self._scan_metadata(all_files)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/tsfile/tsfile.py", line 304, in _scan_metadata
                  from tsfile.constants import TIME_COLUMN, ColumnCategory
              ModuleNotFoundError: No module named 'tsfile'
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/split_names.py", line 66, in compute_split_names_from_streaming_response
                  for split in get_dataset_split_names(
                               ^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/inspect.py", line 340, in get_dataset_split_names
                  info = get_dataset_config_info(
                         ^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/inspect.py", line 291, in get_dataset_config_info
                  raise SplitsNotFoundError("The split names could not be parsed from the dataset config.") from err
              datasets.inspect.SplitsNotFoundError: The split names could not be parsed from the dataset config.

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Time-Series Data (Teeny-Tiny Castle) — TsFile

Daily product sales time series, converted to Apache TsFile format from the original parquet dataset.

Original dataset: AiresPucrs/time-series-data
Part of: Teeny-Tiny Castle — open-source educational tools for AI Ethics and Safety research
License: apache-2.0
Coverage: 2020-01-01 → 2023-01-02, daily granularity (1098 days)
Scale: 1098 rows · 1 product (chocolate)

About the original dataset

The source is a single parquet file (data/train-*.parquet), 1098 rows × 3 columns. It records daily sales for a single product.

Column	Meaning	dtype	range
`dates`	calendar day, `YYYY-MM-DD`	string	2020-01-01 .. 2023-01-02 (1098 distinct)
`product_id`	product name	string	single value: `chocolate`
`sales`	units sold that day	float64	0 / 103.35 / 466

(product_id, dates) is unique — exactly one row per product per day, no duplicates, no nulls.

How it is stored in this TsFile

One table airespucrs_sales. The product (product_id) is the device/series identity; sales is the measurement over time.

TsFile column	Category	Type	From	Notes
`Time`	TIME	INT64	`dates`	epoch milliseconds (UTC midnight)
`product_id`	TAG	STRING	`product_id`	series identity (device dimension)
`sales`	FIELD	DOUBLE	`sales`	units sold, values unchanged

Modifications made during conversion

Every change is listed below; no other transformation was performed.

Format: parquet → TsFile (single file airespucrs_sales.tsfile).
product_id set as a TAG column (the device/series dimension). The source has a single product, so the TsFile contains one time-series; the tag is kept so the product identity is preserved and the layout generalises to multiple products.
dates → Time: the YYYY-MM-DD string is parsed to INT64 epoch milliseconds (UTC midnight) to form a real time axis.
Original dates string column is dropped. Its information is fully preserved in Time (lossless), so it is not stored as a separate field.
sales precision is preserved: stored as DOUBLE (the source is float64); values are copied verbatim, nothing rounded or rescaled.

Not changed / not dropped

Apart from the dates string column (item 4, losslessly folded into Time), nothing is removed. product_id and sales are retained and the row count is preserved exactly.

Verification

Round-trip checked with the TsFile Python SDK:

row count: TsFile read-back 1,098 == 1,098 staged Parquet ✓
product_id reported as category=TAG; sales as FIELD ✓
sales stored as DOUBLE (source float64) ✓

Usage

from tsfile import TsFileReader
reader = TsFileReader("airespucrs_sales.tsfile")
schemas = reader.get_all_table_schemas()          # table: "airespucrs_sales"
# query field/tag columns; Time is added automatically by the reader

Downloads last month: 62

Total file size:

8.34 kB