Metadata-Version: 2.4
Name: tempo-ql
Version: 0.1.3
Summary: Toolkit and interactive widget for querying time-series healthcare data
Author-email: Venkatesh Sivaraman <venkats@cmu.edu>
License: MIT License
        
        Copyright (c) 2025 CMU Data Interaction Group
        
        Permission is hereby granted, free of charge, to any person obtaining a copy
        of this software and associated documentation files (the "Software"), to deal
        in the Software without restriction, including without limitation the rights
        to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
        copies of the Software, and to permit persons to whom the Software is
        furnished to do so, subject to the following conditions:
        
        The above copyright notice and this permission notice shall be included in all
        copies or substantial portions of the Software.
        
        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
        IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
        FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
        AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
        LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
        OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
        SOFTWARE.
Project-URL: Homepage, https://github.com/cmudig/tempo-ql
Keywords: AI,machine learning,healthcare,EHR,OMOP,MEDS,data science
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numba>=0.59.0
Requires-Dist: numpy>=1.26.0
Requires-Dist: pandas>=2.0.0
Requires-Dist: scikit-learn>=1.3.2
Requires-Dist: tqdm>=4.66.1
Requires-Dist: lark>=1.1.8
Requires-Dist: pyarrow>=13.0.0
Requires-Dist: ipywidgets
Requires-Dist: anywidget
Requires-Dist: scipy>=1.11.3
Requires-Dist: sqlalchemy>=2.0.31
Requires-Dist: duckdb-engine
Requires-Dist: google-genai
Dynamic: license-file

# TempoQL: Standardized Temporal Queries for ML in Healthcare

## Quickstart

You may first want to create a conda environment to install packages in. *Use a Python version between 3.10 and 3.14 for compatibility with TempoQL's dependencies.* Then run:

```bash
pip install tempo-ql
```

*We have had issues in the past running the JupyterLab widget with virtualenv - therefore we recommend using conda.*

## Online Examples

There are two examples on Google Colab to show you how TempoQL can be used, both involving the MIMIC-IV dataset but in different data formats:

* [MIMIC-IV Demo Data in OMOP Common Data Model](https://colab.research.google.com/drive/1ttyFJuPUn37b4zXq0PzDxRbL1QBf3x0p?usp=sharing) - uses only non-credentialed data
* [MIMIC-IV Full Dataset on BigQuery](https://colab.research.google.com/drive/1t2lRa0A0ojBjWhTSY7m7RKiXAZDTOCAp?usp=sharing) - requires credentialed PhysioNet access and a Google Cloud Platform account. Follow the instructions on [this page](https://physionet.org/content/mimiciv/3.1/) to get access.

## Example Usage

The `demo.ipynb` and `demo_mimiciv_full.ipynb` notebooks in the repo (or the Colab examples above) shows how to use the query language using MIMIC-IV in OMOP format. You can run these to explore how TempoQL enables simple, readable and precise queries on EHR data.

You will need a dataset and a dataset specification to start using TempoQL. Then, you can import TempoQL and use it in your Python code like this:

```python
from tempo_ql import QueryEngine, GenericDataset, formats

db_specification = formats.omop() # also available: mimiciv(), eicu()
sql_connection_string = "bigquery://my-project" # or "duckdb://my_local_db", etc.
dataset = GenericDataset(sql_connection_string, db_specification)

query_engine = QueryEngine(dataset)
# see demo.ipynb for further options, such as configuring a variable store

query_engine.list_data_elements(scope="Measurement") # returns a dataframe of Measurement concepts

query_engine.query("{Temperature Celsius; scope = Measurement}") # retrieves temperature measurements
```

You can access the interactive query authoring environment in a Jupyter notebook (or VSCode IPython notebook) like so:

```python
query_engine.interactive(file_path=..., api_key=...)
```

Both `file_path` and `api_key` are optional. `file_path` allows you to read and
write queries from a local JSON file, enabling you to persist the queries that you
create in the interactive session. `api_key` can be a Gemini API key allowing you
to use LLMs to author, update, explain, and debug queries.

## Dev Notes

**For local install:** clone the repo, cd into it and run `pip install -e .`.

**Running the dev server:** Make sure you have NodeJS version 20 or later. `cd` into the `client` directory, run `npm install`, then `npm run dev`. Then in your call to `QueryEngine.interactive`, set `dev=True`. Now when you change the frontend source code, the widget will automatically update.

If the Vite dev server stops working after you make some changes (it may show a JavaScript error like 'failed to load model'), check that any imports of TypeScript types are prefixed with the word `type`.
