Skip to content

README.md

Latest commit

 

History

History
41 lines (31 loc) · 796 Bytes

README.md

File metadata and controls

41 lines (31 loc) · 796 Bytes

SensorTSLM Core Framework

Dataset-agnostic captioning pipeline for sensor time-series data.

Setup

Set the dataset path before running:

export MHC_DATASET_DIR="../hf-daily_max-nonwear=50"

Usage

python captionizer.py

Export To TimeF

from pathlib import Path

from captionizer import Captionizer
from exporters.timef_export import TimeFExportConfig, export_caption_result

result, _ = captionizer.run(max_rows=5)
root = export_caption_result(
    result,
    TimeFExportConfig(
        output_root=Path("exports"),
        dataset_id="mhc_caption_runs",
        sampling_period=1,
        timestamp_unit=1,
        unit_sampling_rate="1 / minute",
        unit_timestamp="minute",
        time_column_name="time_minute",
    ),
)
print(root)