README.md

---
license: mit
task_categories:
- time-series-forecasting
- tabular-classification
tags:
- tsfile
- timeseries
- time-series
- finance
- econophysics
- algorithmic-information-theory
- structural-breaks
- anomaly-detection
- format:tsfile
pretty_name: Computational Phase Transitions (TsFile)
size_categories:
- 10M<n<100M
---

# Financial Structural Breaks & Regime Detection Benchmark (TsFile)

Apache TsFile version of [`algoplexity/computational-phase-transitions-data`](https://huggingface.co/datasets/algoplexity/computational-phase-transitions-data).

## Overview

Large-scale collection of non-stationary, continuous financial time series as an
immutable data artifact for the Algoplexity research program into Algorithmic
Information Dynamics (AID) in financial markets. Each `id` is a trajectory — an
ordered sequence of `(time, value)` pairs under a time-varying control parameter
`period` — with a per-trajectory boolean label `structural_breakpoint` (whether
the trajectory exhibits a structural/phase break). The source is split into
`train` and a reduced `test` set; that split is preserved. Because the train
split is very large (23.7M rows), it is stored as six sharded files
(`*_train.tsfile`, `*_train_1.tsfile`, …, `*_train_5.tsfile`) that together form
one logical table.

## Schema (TsFile structure)

- **Time** (INT64, milliseconds) — the step index within each trajectory (not
  wall-clock time).
- **id** (TAG) — the trajectory identifier.
- **structural_breakpoint** (TAG) — the per-trajectory label (True/False).
- **period** (FIELD, INT64) — the time-varying control parameter.
- **value** (FIELD, FLOAT) — the trajectory value.

## Usage

Install the Apache TsFile Python SDK (`pip install tsfile`) and read a converted file:

```python
from pathlib import Path
from tsfile import TsFileReader

path = Path("computational_phase_transitions_data_test.tsfile")
with TsFileReader(str(path)) as reader:
    schemas = reader.get_all_table_schemas()
    print("tables:", list(schemas))
    table_name = next(iter(schemas))
    table = schemas[table_name]
    columns = [column.get_column_name() for column in table.get_columns()]
    print("columns:", columns)
    field_names = [
        column.get_column_name()
        for column in table.get_columns()
        if column.get_column_name() not in {"Time", "time"}
    ]
    if field_names:
        with reader.query_table(table_name, field_names[:3], batch_size=1024) as result:
            batch = result.read_arrow_batch()
            if batch is not None:
                print(batch.to_pandas().head())
```

## Source & license

- Original dataset: https://huggingface.co/datasets/algoplexity/computational-phase-transitions-data
- Author / publisher: algoplexity
- License: MIT