remove upset diff plot

The upsetplot package is incompatible with numpy v2 and pandas v2, both
which are required for lyscripts v1.

To reproduce the upset diff plot, one has therefore to go back in the
git history a little.

Author: rmnldwg

2025-hvh-oropharynx/figures/factors_diff.png

@@ -167,8 +167,8 @@ stages:
       size: 21351
     - path: scripts/render.py
       hash: md5
-      md5: c9d7f0af89624327880cc89d3ecbc291
-      size: 1986
+      md5: 097a0bf7d0a39981ee0b4f403bd534fc
+      size: 2096
     outs:
     - path: 2021-clb-oropharynx/README.md
       hash: md5
@@ -192,8 +192,8 @@ stages:
       size: 28008
     - path: scripts/render.py
       hash: md5
-      md5: c9d7f0af89624327880cc89d3ecbc291
-      size: 1986
+      md5: 097a0bf7d0a39981ee0b4f403bd534fc
+      size: 2096
     outs:
     - path: 2023-clb-multisite/README.md
       hash: md5
@@ -217,8 +217,8 @@ stages:
       size: 52984
     - path: scripts/render.py
       hash: md5
-      md5: c9d7f0af89624327880cc89d3ecbc291
-      size: 1986
+      md5: 097a0bf7d0a39981ee0b4f403bd534fc
+      size: 2096
     outs:
     - path: 2023-isb-multisite/README.md
       hash: md5
@@ -233,8 +233,8 @@ stages:
       size: 166966
     - path: scripts/t_category.py
       hash: md5
-      md5: 3e3c62e6daa7a90f7f4e40d90d00b0f4
-      size: 1622
+      md5: cf119e4a12ccd9a42cd55ac3bc863c4b
+      size: 1690
   plot-t-category@2021-clb-oropharynx:
     cmd: "python scripts/t_category.py 2021-clb-oropharynx/data.csv\n"
     deps:
@@ -244,8 +244,8 @@ stages:
       size: 92744
     - path: scripts/t_category.py
       hash: md5
-      md5: 3e3c62e6daa7a90f7f4e40d90d00b0f4
-      size: 1622
+      md5: cf119e4a12ccd9a42cd55ac3bc863c4b
+      size: 1690
   plot-t-category@2023-clb-multisite:
     cmd: "python scripts/t_category.py 2023-clb-multisite/data.csv\n"
     deps:
@@ -255,8 +255,8 @@ stages:
       size: 111721
     - path: scripts/t_category.py
       hash: md5
-      md5: 3e3c62e6daa7a90f7f4e40d90d00b0f4
-      size: 1622
+      md5: cf119e4a12ccd9a42cd55ac3bc863c4b
+      size: 1690
   plot-t-category@2023-isb-multisite:
     cmd: "python scripts/t_category.py 2023-isb-multisite/data.csv\n"
     deps:
@@ -266,8 +266,8 @@ stages:
       size: 204462
     - path: scripts/t_category.py
       hash: md5
-      md5: 3e3c62e6daa7a90f7f4e40d90d00b0f4
-      size: 1622
+      md5: cf119e4a12ccd9a42cd55ac3bc863c4b
+      size: 1690
   plot-t-category@2025-hvh-oropharynx:
     cmd: "python scripts/t_category.py 2025-hvh-oropharynx/data.csv\n"
     deps:
@@ -277,8 +277,8 @@ stages:
       size: 82567
     - path: scripts/t_category.py
       hash: md5
-      md5: 3e3c62e6daa7a90f7f4e40d90d00b0f4
-      size: 1622
+      md5: cf119e4a12ccd9a42cd55ac3bc863c4b
+      size: 1690
   plot-age-and-sex@2021-usz-oropharynx:
     cmd: "python scripts/age_and_sex.py 2021-usz-oropharynx/data.csv\n"
     deps:
@@ -288,8 +288,8 @@ stages:
       size: 166966
     - path: scripts/age_and_sex.py
       hash: md5
-      md5: fba372b310fd04a38564b4f9dab808a9
-      size: 3160
+      md5: 9fc68d3a1f95e03d7262ef3225b28a32
+      size: 3252
   plot-age-and-sex@2021-clb-oropharynx:
     cmd: "python scripts/age_and_sex.py 2021-clb-oropharynx/data.csv\n"
     deps:
@@ -299,8 +299,8 @@ stages:
       size: 92744
     - path: scripts/age_and_sex.py
       hash: md5
-      md5: fba372b310fd04a38564b4f9dab808a9
-      size: 3160
+      md5: 9fc68d3a1f95e03d7262ef3225b28a32
+      size: 3252
   plot-age-and-sex@2023-clb-multisite:
     cmd: "python scripts/age_and_sex.py 2023-clb-multisite/data.csv\n"
     deps:
@@ -310,8 +310,8 @@ stages:
       size: 111721
     - path: scripts/age_and_sex.py
       hash: md5
-      md5: fba372b310fd04a38564b4f9dab808a9
-      size: 3160
+      md5: 9fc68d3a1f95e03d7262ef3225b28a32
+      size: 3252
   plot-age-and-sex@2023-isb-multisite:
     cmd: "python scripts/age_and_sex.py 2023-isb-multisite/data.csv\n"
     deps:
@@ -321,8 +321,8 @@ stages:
       size: 204462
     - path: scripts/age_and_sex.py
       hash: md5
-      md5: fba372b310fd04a38564b4f9dab808a9
-      size: 3160
+      md5: 9fc68d3a1f95e03d7262ef3225b28a32
+      size: 3252
   plot-age-and-sex@2025-hvh-oropharynx:
     cmd: "python scripts/age_and_sex.py 2025-hvh-oropharynx/data.csv\n"
     deps:
@@ -332,8 +332,8 @@ stages:
       size: 82567
     - path: scripts/age_and_sex.py
       hash: md5
-      md5: fba372b310fd04a38564b4f9dab808a9
-      size: 3160
+      md5: 9fc68d3a1f95e03d7262ef3225b28a32
+      size: 3252
   plot-subsite@2021-usz-oropharynx:
     cmd: "python scripts/subsite.py --data 2021-usz-oropharynx/data.csv\n"
     deps:
@@ -398,8 +398,8 @@ stages:
       size: 92744
     - path: scripts/bar_plot.py
       hash: md5
-      md5: 094dfad6e1e55f49472dac08829ef089
-      size: 2577
+      md5: 16bb00580dbb46db7f52f6f9a3f59276
+      size: 2628
   bar-plot@2023-clb-multisite:
     cmd: python scripts/bar_plot.py 2023-clb-multisite/data.csv
     deps:
@@ -409,8 +409,8 @@ stages:
       size: 111721
     - path: scripts/bar_plot.py
       hash: md5
-      md5: 094dfad6e1e55f49472dac08829ef089
-      size: 2577
+      md5: 16bb00580dbb46db7f52f6f9a3f59276
+      size: 2628
   bar-plot@2023-isb-multisite:
     cmd: python scripts/bar_plot.py 2023-isb-multisite/data.csv
     deps:
@@ -420,5 +420,22 @@ stages:
       size: 204462
     - path: scripts/bar_plot.py
       hash: md5
-      md5: 094dfad6e1e55f49472dac08829ef089
-      size: 2577
+      md5: 16bb00580dbb46db7f52f6f9a3f59276
+      size: 2628
+  factors-diff@2021-usz-oropharynx-2025-hvh-oropharynx:
+    cmd: "python scripts/factors_diff.py --first-dataset 2021-usz-oropharynx --second-dataset
+      2025-hvh-oropharynx --commit ee15c4cabdd160a10783fcf85ffe1cbfeb5c4826 --repo
+      rmnldwg/lydata.private\n"
+    deps:
+    - path: 2021-usz-oropharynx/data.csv
+      hash: md5
+      md5: a9e4043ed6a273c609fa16523bcae455
+      size: 166966
+    - path: 2025-hvh-oropharynx/data.csv
+      hash: md5
+      md5: 7baa8021efbff0126fcf297e52106168
+      size: 82567
+    - path: scripts/factors_diff.py
+      hash: md5
+      md5: 182812f0fe01deba4fd1669153b88ae7
+      size: 3775

dvc.lock

@@ -183,20 +183,6 @@ stages:
       cmd:
         python scripts/bar_plot.py ${item}/data.csv
 
-  upset-diff:
-    matrix:
-      first: [2021-usz-oropharynx]
-      second: [2025-hvh-oropharynx]
-    cmd: >
-      python scripts/upset_diff.py
-      --first-dataset ${item.first}
-      --second-dataset ${item.second}
-      --commit eb8a38ec56269d8c8f23d7155ecaad93f399b5f4
-    deps:
-      - ${item.first}/data.csv
-      - ${item.second}/data.csv
-      - scripts/upset_diff.py
-
   factors-diff:
     matrix:
       first: [2021-usz-oropharynx]
@@ -205,7 +191,8 @@ stages:
       python scripts/factors_diff.py
       --first-dataset ${item.first}
       --second-dataset ${item.second}
-      --commit eb8a38ec56269d8c8f23d7155ecaad93f399b5f4
+      --commit ee15c4cabdd160a10783fcf85ffe1cbfeb5c4826
+      --repo rmnldwg/lydata.private
     deps:
       - ${item.first}/data.csv
       - ${item.second}/data.csv

dvc.yaml

@@ -1,5 +1,4 @@
-
-numpy == 2.2.2
+numpy == 2.2.3
 pandas == 2.2.3
 matplotlib == 3.10.0
 dvc == 3.59.0
@@ -9,4 +8,3 @@ icd10-cm == 0.0.5
 lyscripts == 1.0.0.a5
 lazydocs == 0.4.8
 tueplots == 0.0.17
-upsetplot == 0.9.0

requirements.in

@@ -401,7 +401,7 @@ numexpr==2.10.2
     # via
     #   blosc2
     #   tables
-numpy==2.2.2
+numpy==1.26.4
     # via
     #   -r requirements.in
     #   blosc2
@@ -444,15 +444,15 @@ packaging==24.2
     #   nbconvert
     #   pandera
     #   tables
-pandas==2.2.3
+pandas==1.5.3
     # via
     #   -r requirements.in
     #   lydata
     #   lymph-model
     #   lyscripts
     #   pandera
     #   upsetplot
-pandera==0.23.0
+pandera==0.22.1
     # via lydata
 pandocfilters==1.5.1
     # via nbconvert
@@ -727,7 +727,6 @@ tzdata==2025.1
     # via
     #   celery
     #   kombu
-    #   pandas
 upsetplot==0.9.0
     # via -r requirements.in
 uri-template==1.3.0

requirements.txt

@@ -1,19 +1,58 @@
 """Plot the difference of clinicopathological factors in two datasets as a bar plot."""
 
+import argparse
 from pathlib import Path
 
 import matplotlib.pyplot as plt
 from lyscripts.plots import COLORS
 from shared import MPLSTYLE
 from tueplots import figsizes, fontsizes
-from upset_diff import get_parser, kwargs_from_option
 
 import lydata
 from lydata import C
 
 OUTPUT_NAME = Path(__file__).with_suffix(".png").name
 
 
+def get_parser() -> argparse.ArgumentParser:
+    """Return the argument parser."""
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument(
+        "--repo",
+        type=str,
+        help="The repository from which to load the datasets.",
+        default="rmnldwg/lydata",
+    )
+    parser.add_argument(
+        "--commit",
+        type=str,
+        help="The commit hash at which to compare the datasets.",
+        default="5b85184ecece020f509ab0c9f05aa5c81257ffd3",
+    )
+    parser.add_argument(
+        "--first-dataset",
+        type=str,
+        default="2021-usz-oropharynx",
+    )
+    parser.add_argument(
+        "--second-dataset",
+        type=str,
+        default="2025-hvh-oropharynx",
+    )
+    return parser
+
+
+def kwargs_from_option(option: str) -> dict:
+    """Return the load_kwargs for the given option."""
+    year, institution, subsite = option.split("-")
+    return {
+        "year": int(year),
+        "institution": institution,
+        "subsite": subsite,
+        "use_github": True,
+    }
+
+
 def create_ax() -> plt.Axes:
     """Create the axis for the plot."""
     plt.style.use(MPLSTYLE)
@@ -28,7 +67,7 @@ def main() -> None:
     args = get_parser().parse_args()
 
     load_kwargs = {
-        "repo": "rmnldwg/lydata",
+        "repo_name": args.repo,
         "ref": args.commit,
     }