Merge pull request #29 from mmcdermott/reusable_interface

Making the interface re-usable and importable.

README.md

@@ -420,3 +420,14 @@ To use either of these, you need to install additional optional dependencies:
    files. Similar to task configuration files, but for models.
 2. Figure out how to ensure that each pre-processing step reads from the right prior files. Likely need some
    kind of a "prior stage name" config variable.
+
+## Notes:
+
+You can overwrite the `stages` parameter on the command line to run a dynamic pipeline with just a subset of
+options (the `--cfg job --resolve` is just to make hydra show the induced, resolved config instead of trying
+to run anything):
+
+```bash
+MEDS_polars_functions on  reusable_interface [$⇡] is 󰏗 v0.0.1 via  v3.12.4 via  MEDS_fns
+❯ ./src/MEDS_polars_functions/scripts/preprocessing/normalize.py input_dir=foo cohort_dir=bar 'stages=["normalize", "tensorize"]' --cfg job --resolve
+```

pyproject.toml

@@ -1,5 +1,5 @@
 [build-system]
-requires = ["setuptools>=61.0"]
+requires = ["setuptools>=61.0", "setuptools-scm>=8.0", "wheel"]
 build-backend = "setuptools.build_meta"
 
 [project]
@@ -16,7 +16,7 @@ classifiers = [
     "License :: OSI Approved :: MIT License",
     "Operating System :: OS Independent",
 ]
-dependencies = ["polars", "pyarrow", "nested_ragged_tensors", "loguru", "hydra-core", "numpy"]
+dependencies = ["polars>=1.1.0", "pyarrow", "nested_ragged_tensors", "loguru", "hydra-core", "numpy"]
 
 [project.optional-dependencies]
 examples = ["rootutils"]
@@ -25,6 +25,13 @@ tests = ["pytest", "pytest-cov", "rootutils"]
 local_parallelism = ["hydra-joblib-launcher"]
 slurm_parallelism = ["hydra-submitit-launcher"]
 
+[project.scripts]
+MEDS_extract-split_and_shard_patients = "MEDS_polars_functions.extraction.split_and_shard_patients:main"
+MEDS_extract-shard_events = "MEDS_polars_functions.extraction.shard_events:main"
+MEDS_extract-convert_to_sharded_events = "MEDS_polars_functions.extraction.convert_to_sharded_events:main"
+MEDS_extract-merge_to_MEDS_cohort = "MEDS_polars_functions.extraction.merge_to_MEDS_cohort:main"
+MEDS_transform-aggregate_code_metadata = "MEDS_polars_functions.aggregate_code_metadata:main"
+
 [project.urls]
 Homepage = "https://github.com/mmcdermott/MEDS_polars_functions"
 Issues = "https://github.com/mmcdermott/MEDS_polars_functions/issues"