Merge branch 'main' into clean

mmcdermott · Jun 12, 2024 · 109d153 · 109d153
2 parents d7127e8 + 1009f70
commit 109d153
Show file tree

Hide file tree

Showing 25 changed files with 401 additions and 22 deletions.
diff --git a/README.md b/README.md
@@ -3,14 +3,16 @@
 This repository provides utilities and scripts to run limited automatic tabular ML pipelines for generic MEDS
 datasets.
 
-
 # Installation
 
 **Pip Install**
+
 ```bash
 pip install meds-tab
 ```
+
 **Local Install**
+
 ```
 # clone the git repo
 pip install .
@@ -39,7 +41,8 @@ For an end to end example over MIMIC-IV, see the [companion repository](https://
 ### Core CLI Scripts Overview
 
 1. **`meds-tab-describe`**: This command processes MEDS data shards to compute the frequencies of different code-types
-	- time-series codes (codes with timestamps)
+
+   - time-series codes (codes with timestamps)
    - time-series numerical values (codes with timestamps and numerical values)
    - static codes (codes without timestamps)
    - static numerical codes (codes without timestamps but with numerical values).
@@ -54,6 +57,7 @@ For an end to end example over MIMIC-IV, see the [companion repository](https://
 
 6. **`meds-tab-xgboost-sweep`**: Conducts an Optuna hyperparameter sweep to optimize over `window_sizes`, `aggregations`, and `min_code_inclusion_frequency`, aiming to enhance model performance and adaptability.
 
+# How does MEDS-Tab Work?
 
 #### What do you mean "tabular pipelines"? Isn't _all_ structured EHR data already tabular?
 
@@ -66,4 +70,14 @@ satisfy the (1) "single row single instance", (2) "consistent set of columns", a
 Thus, in this pipeline, when we say we will produce a "tabular" view of MEDS data, we mean a dataset that can
 realize these constraints, which will explicitly involve summarizing the patient data over various historical
 or future windows in time to produce a single row per patient with a consistent, logical set of columns
-(though there may still be missingness).
+(though there may still be missingness).
+
+## Implementation Improvements
+
+# Computational Performance vs. Existing Pipelines
+
+# XGBoost Performance
+
+## XGBoost Model Performance on MIMIC-IV
+
+## XGBoost Model Performance on Philips eICU
diff --git a/docs/source/computational-performance.rst b/docs/source/computational-performance.rst
@@ -0,0 +1,7 @@
+Computational Performance vs. Existing Pipelines
+================================================
+
+.. include:: ../../README.md
+   :parser: markdown
+   :start-after: Computational Performance vs. Existing Pipelines
+   :end-before: XGBoost Model Performance on MIMIC-IV
diff --git a/docs/source/conf.py b/docs/source/conf.py
@@ -9,16 +9,17 @@
 # -- Project information -----------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#project-information
 
-project = "MEDS-TAB"
+project = "MEDS-Tab"
 copyright = "2024, Matthew McDermott, Nassim Oufattole, Teya Bergamaschi"
 author = "Matthew McDermott, Nassim Oufattole, Teya Bergamaschi"
-release = "0.1.0"
-version = "0.1.0"
+release = "0.0.1"
+version = "0.0.1"
 
 # -- General configuration ---------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#general-configuration
 
 sys.path.insert(0, os.path.abspath("../.."))
+
 extensions = [
     "sphinx.ext.duration",
     "sphinx.ext.doctest",
@@ -60,8 +61,8 @@
 html_static_path = ["_static"]
 
 
-html_title = f"NEDS-TAB v{version} Documentation"
-html_short_title = "MEDS-TAB Documentation"
+html_title = f"MEDS-Tab v{version} Documentation"
+html_short_title = "MEDS-Tab Documentation"
 
 # html_logo = "query-512.png"
 # html_favicon = "query-16.ico"
@@ -70,7 +71,7 @@
 
 html_theme_options = {
     "dark_mode_code_blocks": False,
-    # "nav_title": "MEDS-TAB",
+    # "nav_title": "MEDS-Tab",
     # "palette": {"primary": "green", "accent": "green"},
     # "repo_url": "https://github.com/mmcdermott/MEDS_Tabular_AutoML",
     # "repo_name": "MEDS_Tabular_AutoML",

diff --git a/docs/source/generated/src.MEDS_tabular_automl.configs.rst b/docs/source/generated/src.MEDS_tabular_automl.configs.rst
@@ -0,0 +1,30 @@
+src.MEDS\_tabular\_automl.configs
+=================================
+
+.. automodule:: src.MEDS_tabular_automl.configs
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+.. rubric:: Modules
+
+.. autosummary::
+   :toctree:
+   :recursive:
+
+   src.MEDS_tabular_automl.configs.tabularization
diff --git a/docs/source/generated/src.MEDS_tabular_automl.configs.tabularization.rst b/docs/source/generated/src.MEDS_tabular_automl.configs.tabularization.rst
@@ -0,0 +1,4 @@
+src.MEDS\_tabular\_automl.configs.tabularization
+================================================
+
+.. automodule:: src.MEDS_tabular_automl.configs.tabularization
diff --git a/docs/source/generated/src.MEDS_tabular_automl.describe_codes.rst b/docs/source/generated/src.MEDS_tabular_automl.describe_codes.rst
@@ -0,0 +1,23 @@
+src.MEDS\_tabular\_automl.describe\_codes
+=========================================
+
+.. automodule:: src.MEDS_tabular_automl.describe_codes
+
+
+
+
+
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      clear_code_aggregation_suffix
+      compute_feature_frequencies
+      convert_to_df
+      convert_to_freq_dict
+      filter_parquet
+      filter_to_codes
+      get_feature_columns
+      get_feature_freqs
diff --git a/docs/source/generated/src.MEDS_tabular_automl.file_name.rst b/docs/source/generated/src.MEDS_tabular_automl.file_name.rst
@@ -0,0 +1,18 @@
+src.MEDS\_tabular\_automl.file\_name
+====================================
+
+.. automodule:: src.MEDS_tabular_automl.file_name
+
+
+
+
+
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      get_model_files
+      get_task_specific_path
+      list_subdir_files
diff --git a/docs/source/generated/src.MEDS_tabular_automl.generate_static_features.rst b/docs/source/generated/src.MEDS_tabular_automl.generate_static_features.rst
@@ -0,0 +1,19 @@
+src.MEDS\_tabular\_automl.generate\_static\_features
+====================================================
+
+.. automodule:: src.MEDS_tabular_automl.generate_static_features
+
+
+
+
+
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      convert_to_matrix
+      get_flat_static_rep
+      get_sparse_static_rep
+      summarize_static_measurements
diff --git a/docs/source/generated/src.MEDS_tabular_automl.generate_summarized_reps.rst b/docs/source/generated/src.MEDS_tabular_automl.generate_summarized_reps.rst
@@ -0,0 +1,20 @@
+src.MEDS\_tabular\_automl.generate\_summarized\_reps
+====================================================
+
+.. automodule:: src.MEDS_tabular_automl.generate_summarized_reps
+
+
+
+
+
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      aggregate_matrix
+      compute_agg
+      generate_summary
+      get_rolling_window_indicies
+      sparse_aggregate
diff --git a/docs/source/generated/src.MEDS_tabular_automl.generate_ts_features.rst b/docs/source/generated/src.MEDS_tabular_automl.generate_ts_features.rst
@@ -0,0 +1,20 @@
+src.MEDS\_tabular\_automl.generate\_ts\_features
+================================================
+
+.. automodule:: src.MEDS_tabular_automl.generate_ts_features
+
+
+
+
+
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      feature_name_to_code
+      get_flat_ts_rep
+      get_long_code_df
+      get_long_value_df
+      summarize_dynamic_measurements
diff --git a/docs/source/generated/src.MEDS_tabular_automl.mapper.rst b/docs/source/generated/src.MEDS_tabular_automl.mapper.rst
@@ -0,0 +1,18 @@
+src.MEDS\_tabular\_automl.mapper
+================================
+
+.. automodule:: src.MEDS_tabular_automl.mapper
+
+
+
+
+
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      get_earliest_lock
+      register_lock
+      wrap
diff --git a/docs/source/generated/src.MEDS_tabular_automl.rst b/docs/source/generated/src.MEDS_tabular_automl.rst
@@ -2,3 +2,37 @@ src.MEDS\_tabular\_automl
 =========================
 
 .. automodule:: src.MEDS_tabular_automl
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+.. rubric:: Modules
+
+.. autosummary::
+   :toctree:
+   :recursive:
+
+   src.MEDS_tabular_automl.configs
+   src.MEDS_tabular_automl.describe_codes
+   src.MEDS_tabular_automl.file_name
+   src.MEDS_tabular_automl.generate_static_features
+   src.MEDS_tabular_automl.generate_summarized_reps
+   src.MEDS_tabular_automl.generate_ts_features
+   src.MEDS_tabular_automl.mapper
+   src.MEDS_tabular_automl.scripts
+   src.MEDS_tabular_automl.utils
diff --git a/docs/source/generated/src.MEDS_tabular_automl.scripts.cache_task.rst b/docs/source/generated/src.MEDS_tabular_automl.scripts.cache_task.rst
@@ -0,0 +1,17 @@
+src.MEDS\_tabular\_automl.scripts.cache\_task
+=============================================
+
+.. automodule:: src.MEDS_tabular_automl.scripts.cache_task
+
+
+
+
+
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      generate_row_cached_matrix
+      main
diff --git a/docs/source/generated/src.MEDS_tabular_automl.scripts.describe_codes.rst b/docs/source/generated/src.MEDS_tabular_automl.scripts.describe_codes.rst
@@ -0,0 +1,16 @@
+src.MEDS\_tabular\_automl.scripts.describe\_codes
+=================================================
+
+.. automodule:: src.MEDS_tabular_automl.scripts.describe_codes
+
+
+
+
+
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      main
diff --git a/docs/source/generated/src.MEDS_tabular_automl.scripts.launch_xgboost.rst b/docs/source/generated/src.MEDS_tabular_automl.scripts.launch_xgboost.rst
@@ -0,0 +1,27 @@
+src.MEDS\_tabular\_automl.scripts.launch\_xgboost
+=================================================
+
+.. automodule:: src.MEDS_tabular_automl.scripts.launch_xgboost
+
+
+
+
+
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      main
+
+
+
+
+
+   .. rubric:: Classes
+
+   .. autosummary::
+
+      Iterator
+      XGBoostModel
diff --git a/docs/source/generated/src.MEDS_tabular_automl.scripts.rst b/docs/source/generated/src.MEDS_tabular_automl.scripts.rst
@@ -0,0 +1,34 @@
+src.MEDS\_tabular\_automl.scripts
+=================================
+
+.. automodule:: src.MEDS_tabular_automl.scripts
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+.. rubric:: Modules
+
+.. autosummary::
+   :toctree:
+   :recursive:
+
+   src.MEDS_tabular_automl.scripts.cache_task
+   src.MEDS_tabular_automl.scripts.describe_codes
+   src.MEDS_tabular_automl.scripts.launch_xgboost
+   src.MEDS_tabular_automl.scripts.tabularize_static
+   src.MEDS_tabular_automl.scripts.tabularize_time_series