TidyRun
A tool to orchestrate the compute and storage of Python DAGs
Features
Compute Orchestration
TidyRun provides first-class deferred compute primitives for DAG workflows:
- Deferred Primitives: Model work with
Job,ParametrizedJob, and nestedDAG - Dependency-Aware Scheduling: Evaluate DAGs with topological execution and fail-fast behavior
- Execution Modes: Choose
subprocess(default),thread, orprocess - Parallel Evaluation: Run independent nodes with
DAG.evaluate(max_workers=...) - Materialized Plans: Compile reproducible execution plans before running jobs
- Resumable Runs: Re-run materialized plans with
execute_materialized(skip_completed=True) - Pluggable Executors: Use local executors,
SlurmExecutor, orAwsBatchExecutor
Serialization and Storage
TidyRun also includes a comprehensive serialization system for storing and retrieving Python objects:
- Type-Aware Encoding: Automatically selects folder, Parquet, HDF5, JSON, or pickle based on value type
- Lazy Evaluation: Directories deserialize into
LazyDictobjects that load values on-demand - Recursive Concatenation: Aggregate DataFrames across nested structures with
LazyDict.concat()(optionally parallel withmax_workers) - Metadata Sidecars: Each output is tracked with
.tidyrunmetadata files for format versioning and checksums - Checksum Return Value:
serialize(...)returns checksum information (algorithm,digest) for the serialized payload - Extensible Pipeline: Customize encoders or add support for custom types
- Intelligent Fallback: Parquet → HDF5 → JSON → Pickle chain ensures robust serialization
Quick Example:
Compute (DAG execution):
from tidyrun import DAG, Job
def square(x: int) -> int:
return x * x
dag = DAG()
dag["a"] = Job(func=square, kwargs={"x": 3})
# Fast local execution without subprocess spawn overhead
outputs = dag.evaluate("./local_dag", execution_mode="thread", max_workers=4)
print(outputs["a"]) # 9
Serialization and lazy loading:
from tidyrun import serialize, deserialize
import pandas as pd
# Save nested data with smart format selection
serialize({
"metrics": pd.DataFrame({"score": [9]}),
"config": {"lr": 0.001},
}, "./results/exp_001")
# Load with lazy evaluation
results = deserialize("./results/exp_001")
df = results["metrics"] # Loads on access
# Aggregate across nested structures
combined = results.concat(names=["run_id"])
Learn More:
- Quick Start — Local docs workflow and publishing notes
- DAG Guide — Jobs, parametrized jobs, executors, and evaluation modes
- Serialization Guide — Complete API reference, quick reference, and examples
Metadata
Release files for tidyrun 0.0.8
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| tidyrun-0.0.8.tar.gz | 48.8 kB | With provenance | Details for tidyrun-0.0.8.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| tidyrun-0.0.8-py3-none-any.whl | Python 3 | none | any | Details for tidyrun-0.0.8-py3-none-any.whl |
Total release size: 110.3 kB
Release files / tidyrun-0.0.8.tar.gz
| Download URL | tidyrun-0.0.8.tar.gz |
|---|---|
| Size | 48.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
a544101bc8bd667782c761f2bd56b521bf8740513c4856b0e8145405e842889c
|
|
BLAKE2b-256 checksum How to use checksums |
bb4d668b4e45f63b1578977d51b76fdea75e2b6527513e4ebc3398184874376c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 14, 2026.
Transparency logRelease files / tidyrun-0.0.8-py3-none-any.whl
| Download URL | tidyrun-0.0.8-py3-none-any.whl |
|---|---|
| Size | 61.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
bb913fc9a7b83d0dd2721e554d309536f01f1163542167388b13a66d11398d11
|
|
BLAKE2b-256 checksum How to use checksums |
ccf54047723f9952ad16adbfa11f3a08f9e26e1b5b26c059cec42a6e8f4ad0fb
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 14, 2026.
Transparency logSecurity for tidyrun 0.0.8
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
Both files in this release have been uploaded with provenance metadata.
Attestation for tidyrun-0.0.8.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 14, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.