Latest scale up batch (#150)

* update dependencies * Update data export and config * Updated last agendas and query * Updated model * Update Changelog * Set max version of sklearn * Fix leading zero's in subagenda * Update manifest and pyproject to python 3.12 * Update GH actions and pyproject
UMCU-Digital-Health · Dec 12, 2024 · 7518823 · 7518823
1 parent 2ba5e2b
commit 7518823
Show file tree

Hide file tree

Showing 20 changed files with 224 additions and 106 deletions.
diff --git a/.github/workflows/linting.yml b/.github/workflows/linting.yml
@@ -19,10 +19,10 @@ jobs:
     - name: Set up Python
       uses: actions/setup-python@v5
       with:
-        python-version: '3.11'
+        python-version: '3.12'
 
     - name: Install dependencies
-      run: uv sync --only-dev
+      run: uv sync --only-group lint
 
     - name: Analysing the code with ruff
       run: |

diff --git a/.github/workflows/unit_test.yml b/.github/workflows/unit_test.yml
@@ -19,7 +19,7 @@ jobs:
     - name: Set up Python
       uses: actions/setup-python@v5
       with:
-        python-version: '3.11'
+        python-version: '3.12'
 
     - name: Install dependencies
       run: uv sync --all-extras --dev

diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -7,6 +7,15 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [1.5.0] - 2024-12-10
+
+### Changed
+- Upgraded to Python 3.12 and updated all dependencies
+- Retrained model on all clinics
+
+### Added
+- Added last batch of agandas to config
+
 ## [1.4.11] - 2024-12-03
 
 ### Changed

diff --git a/data/raw/poliafspraken_no_show.csv.dvc b/data/raw/poliafspraken_no_show.csv.dvc
@@ -1,5 +1,5 @@
 outs:
-- md5: 31983d8a20d487417a9da7cf024c7ecc
-  size: 487004530
+- md5: e33284ae4e740e652ca682f1d6c78fc5
+  size: 806485792
   path: poliafspraken_no_show.csv
   hash: md5
diff --git a/data/sql/data_export.sql b/data/sql/data_export.sql
@@ -129,7 +129,43 @@ WHERE 1=1
             -- Hoofd Hals oncologie
             'ZH0027', -- B&O KNH oncologie
             -- Functie KNF
-            'ZH0175' -- Functie KNF
+            'ZH0175', -- Functie KNF
+            -- Orthopedie
+            'ZH0210', -- Orthopedie
+            -- Urologie
+            'ZH0324', -- Urologie
+            -- Plastische chirurgie
+            'ZH0213', -- Plastische Chirurgie
+            -- Traumatologie Heelkunde
+            'ZH0343', -- Traumatologie
+            'ZH0117', -- Heelkunde
+            -- Vaatchirurgie
+            'ZH0401', -- Vaatcentrum
+            -- KNO
+            'ZH0178', -- KNO
+            -- Functiecentrum KNO
+            'ZH0104', -- Functiecentrum KNO
+            -- MKA
+            'ZH0026', -- B&O Kaakchirurgie en oncologie
+            'ZH0191', -- Mondziekten / kaakchirugie
+            -- Bijzondere tandheelkunde
+            'ZH0002', -- A&G
+            'ZH0014', -- Bijzondere tandheelkunde
+            -- Oogheelkunde
+            'ZH0208', -- Oogheelkunde
+            'CS0140', -- Functie oogheelkunde
+            -- Anesthesiologie
+            'ZH0010', -- Anesthesiologie
+            -- Pijncentrum
+            'ZH0211', -- Pijnbehandeling
+            -- Genetica
+            'ZH0110', -- Genetica
+            'ZH0160', -- Kind-ontwikkelingsachterstand
+            -- Prikpoli
+            '020434', -- Kind-laboratorium
+            'ZH0414', -- VAS-teamagenda            
+            -- Fertiliteit & gyneacologie
+            'ZH04222' -- Voortplanting & gynaecologie
         )
     AND APP.identifier_system = 'https://metadata.umcutrecht.nl/ids/HixAgendaAfspraak'
     AND APP.created >= '2016-01-01'

diff --git a/data/sql/data_prediction.sql b/data/sql/data_prediction.sql
@@ -151,7 +151,43 @@ WHERE 1=1
         -- Hoofd Hals oncologie
         'ZH0027', -- B&O KNH oncologie
         -- Functie KNF
-        'ZH0175' -- Functie KNF
+        'ZH0175', -- Functie KNF
+        -- Orthopedie
+        'ZH0210', -- Orthopedie
+        -- Urologie
+        'ZH0324', -- Urologie
+        -- Plastische chirurgie
+        'ZH0213', -- Plastische Chirurgie
+        -- Traumatologie Heelkunde
+        'ZH0343', -- Traumatologie
+        'ZH0117', -- Heelkunde
+        -- Vaatchirurgie
+        'ZH0401', -- Vaatcentrum
+        -- KNO
+        'ZH0178', -- KNO
+        -- Functiecentrum KNO
+        'ZH0104', -- Functiecentrum KNO
+        -- MKA
+        'ZH0026', -- B&O Kaakchirurgie en oncologie
+        'ZH0191', -- Mondziekten / kaakchirugie
+        -- Bijzondere tandheelkunde
+        'ZH0002', -- A&G
+        'ZH0014', -- Bijzondere tandheelkunde
+        -- Oogheelkunde
+        'ZH0208', -- Oogheelkunde
+        'CS0140', -- Functie oogheelkunde
+        -- Anesthesiologie
+        'ZH0010', -- Anesthesiologie
+        -- Pijncentrum
+        'ZH0211', -- Pijnbehandeling
+        -- Genetica
+        'ZH0110', -- Genetica
+        'ZH0160', -- Kind-ontwikkelingsachterstand
+        -- Prikpoli
+        '020434', -- Kind-laboratorium
+        'ZH0414', -- VAS-teamagenda            
+        -- Fertiliteit & gyneacologie
+        'ZH04222' -- Voortplanting & gynaecologie
     )
     AND APP.identifier_system = 'https://metadata.umcutrecht.nl/ids/HixAgendaAfspraak'
     AND APP.[created] >= '2015-01-01'
@@ -268,7 +304,43 @@ WHERE 1=1
                 -- Hoofd Hals oncologie
                 'ZH0027', -- B&O KNH oncologie
                 -- Functie KNF
-                'ZH0175' -- Functie KNF
+                'ZH0175', -- Functie KNF
+                -- Orthopedie
+                'ZH0210', -- Orthopedie
+                -- Urologie
+                'ZH0324', -- Urologie
+                -- Plastische chirurgie
+                'ZH0213', -- Plastische Chirurgie
+                -- Traumatologie Heelkunde
+                'ZH0343', -- Traumatologie
+                'ZH0117', -- Heelkunde
+                -- Vaatchirurgie
+                'ZH0401', -- Vaatcentrum
+                -- KNO
+                'ZH0178', -- KNO
+                -- Functiecentrum KNO
+                'ZH0104', -- Functiecentrum KNO
+                -- MKA
+                'ZH0026', -- B&O Kaakchirurgie en oncologie
+                'ZH0191', -- Mondziekten / kaakchirugie
+                -- Bijzondere tandheelkunde
+                'ZH0002', -- A&G
+                'ZH0014', -- Bijzondere tandheelkunde
+                -- Oogheelkunde
+                'ZH0208', -- Oogheelkunde
+                'CS0140', -- Functie oogheelkunde
+                -- Anesthesiologie
+                'ZH0010', -- Anesthesiologie
+                -- Pijncentrum
+                'ZH0211', -- Pijnbehandeling
+                -- Genetica
+                'ZH0110', -- Genetica
+                'ZH0160', -- Kind-ontwikkelingsachterstand
+                -- Prikpoli
+                '020434', -- Kind-laboratorium
+                'ZH0414', -- VAS-teamagenda            
+                -- Fertiliteit & gyneacologie
+                'ZH04222' -- Voortplanting & gynaecologie
             )
             AND APP2.identifier_system = 'https://metadata.umcutrecht.nl/ids/HixAgendaAfspraak'
             AND CONVERT(DATE, APP2.[start]) = @start_date

diff --git a/dvc.lock b/dvc.lock
@@ -5,17 +5,17 @@ stages:
     deps:
     - path: data/processed/featuretable.parquet
       hash: md5
-      md5: 4e754146b5f7ea804ab79691e19038d4
-      size: 100497640
+      md5: 752b4606058923722d7d1538bcebc2e6
+      size: 168033517
     - path: src/noshow/model/train_model.py
       hash: md5
-      md5: 12587629eddb081d94bc997c8d100dd4
-      size: 3713
+      md5: 1bb5e77bbacc1e4201645258a982d9b8
+      size: 3959
     outs:
     - path: output/models/no_show_model_cv.pickle
       hash: md5
-      md5: dfc2275999ae50e74a9ab8459372ddfa
-      size: 1939776
+      md5: 5e3f9c01e77a3d4dac17faf920c83ad7
+      size: 1952898
   feature_building:
     cmd: python src/noshow/features/feature_pipeline.py
     deps:
@@ -24,14 +24,14 @@ stages:
       size: 279455
     - path: data/raw/poliafspraken_no_show.csv
       hash: md5
-      md5: 31983d8a20d487417a9da7cf024c7ecc
-      size: 487004530
+      md5: e33284ae4e740e652ca682f1d6c78fc5
+      size: 806485792
     - path: src/noshow/features/feature_pipeline.py
       hash: md5
-      md5: 71ffb7a162976bde11e0aed72ea19f98
-      size: 2889
+      md5: d6364f3b0671017318366eab08084f0f
+      size: 3137
     outs:
     - path: data/processed/featuretable.parquet
       hash: md5
-      md5: 4e754146b5f7ea804ab79691e19038d4
-      size: 100497640
+      md5: 752b4606058923722d7d1538bcebc2e6
+      size: 168033517
diff --git a/manifest_admin_dash.json b/manifest_admin_dash.json
@@ -6,7 +6,7 @@
     "entrypoint": "run/admin_dash.py"
   },
   "python": {
-    "version": "3.11.10",
+    "version": "3.12.6",
     "package_manager": {
       "name": "pip",
       "version": "23.3.1",

diff --git a/manifest_api.json b/manifest_api.json
@@ -6,7 +6,7 @@
     "entrypoint": "run.app:app"
   },
   "python": {
-    "version": "3.11.10",
+    "version": "3.12.6",
     "package_manager": {
       "name": "pip",
       "version": "23.3.1",

diff --git a/manifest_dash.json b/manifest_dash.json
@@ -6,7 +6,7 @@
     "entrypoint": "run/calling_dash.py"
   },
   "python": {
-    "version": "3.11.10",
+    "version": "3.12.6",
     "package_manager": {
       "name": "pip",
       "version": "23.3.1",

diff --git a/output/dvclive/metrics.json b/output/dvclive/metrics.json
@@ -1,7 +1,7 @@
 {
-    "best_score": 0.7480471226772029,
-    "mean_roc_auc": 0.7480471226772029,
-    "std_roc_auc": 0.005831461640562374,
-    "mean_precision": 0.5541110092343222,
-    "mean_recall": 0.014451354691335214
+    "best_score": 0.7555489720970228,
+    "mean_roc_auc": 0.7555489720970228,
+    "std_roc_auc": 0.0038171109326925416,
+    "mean_precision": 0.5476698499321399,
+    "mean_recall": 0.010038742023701001
 }
diff --git a/output/dvclive/plots/metrics/best_score.tsv b/output/dvclive/plots/metrics/best_score.tsv
@@ -1,2 +1,2 @@
 step	best_score
-0	0.7480471226772029
+0	0.7555489720970228
diff --git a/output/dvclive/plots/metrics/mean_precision.tsv b/output/dvclive/plots/metrics/mean_precision.tsv
@@ -1,2 +1,2 @@
 step	mean_precision
-0	0.5541110092343222
+0	0.5476698499321399
diff --git a/output/dvclive/plots/metrics/mean_recall.tsv b/output/dvclive/plots/metrics/mean_recall.tsv
@@ -1,2 +1,2 @@
 step	mean_recall
-0	0.014451354691335214
+0	0.010038742023701001
diff --git a/output/dvclive/plots/metrics/mean_roc_auc.tsv b/output/dvclive/plots/metrics/mean_roc_auc.tsv
@@ -1,2 +1,2 @@
 step	mean_roc_auc
-0	0.7480471226772029
+0	0.7555489720970228
diff --git a/output/dvclive/plots/metrics/std_roc_auc.tsv b/output/dvclive/plots/metrics/std_roc_auc.tsv
@@ -1,2 +1,2 @@
 step	std_roc_auc
-0	0.005831461640562374
+0	0.0038171109326925416
diff --git a/pyproject.toml b/pyproject.toml
@@ -4,60 +4,53 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "noshow"
-version = "1.4.11"
+version = "1.5.0"
 authors = [
   { name="Ruben Peters", email="[email protected]" },
   { name="Eric Wolters", email="[email protected]" }
 ]
 description = "No-show prediction model to reduce the amount of no-shows at the clinics"
 readme = "README.md"
-requires-python = ">=3.11"
+requires-python = ">=3.12"
 classifiers = [
     "Programming Language :: Python :: 3",
     "Operating System :: OS Independent",
 ]
 dependencies =[
     "pandas~=2.0",
     "matplotlib~=3.7",
-    "numpy~=1.24",
-    "scikit-learn~=1.3",
-    "pyarrow~=14.0",
+    "numpy>=1.24",
+    "scikit-learn>=1.3,<1.7",
+    "pyarrow>=14.0",
     "fastapi~=0.109",
     "dvc~=3.33",
     "dvclive~=3.41",
     "sqlalchemy~=2.0",
     "python-dotenv~=1.0",
     "pymssql~=2.2",
-    "streamlit~=1.30",
+    "streamlit>=1.30",
     "tomli~=2.0",
     "relplot>=1.0",
 ]
 
-[project.optional-dependencies]
-test = ["pytest", "pytest-cov", "pytest-asyncio"]
-dev = ["rsconnect-python", "uvicorn", "nbstripout", "ipykernel"]
-
-[tool.rye]
-dev-dependencies = [
-    "ipykernel>=6.29.2",
-    "pytest>=8.0.1",
-    "nbstripout>=0.7.1",
-    "uvicorn~=0.28.0",
-    "pytest-asyncio~=0.23.6",
-]
-
-[tool.uv]
-dev-dependencies = [
+[dependency-groups]
+dev = [
     "ipykernel>=6.29.2",
     "pytest>=8.0.1",
     "nbstripout>=0.7.1",
-    "uvicorn~=0.28.0",
-    "pytest-asyncio~=0.23.6",
+    "uvicorn>=0.28.0",
+    "pytest-asyncio>=0.23.6",
     "pytest-cov>=5.0.0",
     "alembic>=1.13.3",
+    "rsconnect-python>=1.24.0",
+]
+lint = [
     "ruff>=0.7.1",
 ]
 
+[tool.uv]
+default-groups = ["dev", "lint"]
+
 [tool.isort]
 profile = "black"