From b8f0c8cd610c4289fa603daa59898b55784847fc Mon Sep 17 00:00:00 2001
From: Andre Anjos <andre.dos.anjos@gmail.com>
Date: Thu, 20 Jul 2023 23:34:12 +0200
Subject: [PATCH] [data.montgomery/shenzhen] Simplify documentation maintenance

---
 src/ptbench/data/montgomery/default.py | 23 +++++++++++----
 src/ptbench/data/montgomery/fold_0.py  | 24 ++-------------
 src/ptbench/data/montgomery/fold_1.py  | 22 +-------------
 src/ptbench/data/montgomery/fold_2.py  | 22 +-------------
 src/ptbench/data/montgomery/fold_3.py  | 22 +-------------
 src/ptbench/data/montgomery/fold_4.py  | 22 +-------------
 src/ptbench/data/montgomery/fold_5.py  | 22 +-------------
 src/ptbench/data/montgomery/fold_6.py  | 22 +-------------
 src/ptbench/data/montgomery/fold_7.py  | 22 +-------------
 src/ptbench/data/montgomery/fold_8.py  | 22 +-------------
 src/ptbench/data/montgomery/fold_9.py  | 22 +-------------
 src/ptbench/data/shenzhen/default.py   | 41 ++++++++++++++++----------
 src/ptbench/data/shenzhen/fold_0.py    | 26 +---------------
 src/ptbench/data/shenzhen/fold_1.py    | 26 +---------------
 src/ptbench/data/shenzhen/fold_2.py    | 26 +---------------
 src/ptbench/data/shenzhen/fold_3.py    | 26 +---------------
 src/ptbench/data/shenzhen/fold_4.py    | 26 +---------------
 src/ptbench/data/shenzhen/fold_5.py    | 26 +---------------
 src/ptbench/data/shenzhen/fold_6.py    | 26 +---------------
 src/ptbench/data/shenzhen/fold_7.py    | 26 +---------------
 src/ptbench/data/shenzhen/fold_8.py    | 26 +---------------
 src/ptbench/data/shenzhen/fold_9.py    | 26 +---------------
 22 files changed, 63 insertions(+), 483 deletions(-)

diff --git a/src/ptbench/data/montgomery/default.py b/src/ptbench/data/montgomery/default.py
index 35878b1a..bb57b9a7 100644
--- a/src/ptbench/data/montgomery/default.py
+++ b/src/ptbench/data/montgomery/default.py
@@ -2,16 +2,27 @@
 #
 # SPDX-License-Identifier: GPL-3.0-or-later
 
-"""Montgomery datamodule for TB detection (default protocol)
+"""Montgomery datamodule for TB detection (``default`` protocol)
 
-* See :py:mod:`ptbench.data.montgomery` for more database details.
+The standard digital image database for Tuberculosis was created by the National
+Library of Medicine, Maryland, USA in collaboration with Shenzhen No.3 People’s
+Hospital, Guangdong Medical College, Shenzhen, China. The Chest X-rays are from
 
-This configuration:
+* Database reference: [MONTGOMERY-SHENZHEN-2014]_
+* Original resolution (height x width or width x height): 4020x4892 px or 4892x4020 px
+* This split:
+
+  * Split reference: None
+  * Training samples: ?? of TB and healthy CXR
+  * Validation samples: ?? of TB and healthy CXR
+  * Test samples: ?? of TB and healthy CXR
+
+Data specifications:
 
 * Raw data input (on disk):
 
     * PNG images 8 bit grayscale
-    * resolution: 4020 x 4892 px or 4892 x 4020 px
+    * resolution: fixed to one of the cases above
 
 * Output image:
 
@@ -23,8 +34,8 @@ This configuration:
 
     * Final specifications
 
-        * Grayscale (single channel), 8 bits
-        * Varying resolutions
+        * Grayscale, encoded as a single plane image, 8 bits
+        * Square (4020x4020 px)
 
 
 Protocol ``default``:
diff --git a/src/ptbench/data/montgomery/fold_0.py b/src/ptbench/data/montgomery/fold_0.py
index 4bfc4784..e50d2e30 100644
--- a/src/ptbench/data/montgomery/fold_0.py
+++ b/src/ptbench/data/montgomery/fold_0.py
@@ -2,29 +2,9 @@
 #
 # SPDX-License-Identifier: GPL-3.0-or-later
 
-"""Montgomery datamodule for TB detection (default protocol)
+"""Montgomery datamodule for TB detection (``fold 0`` protocol)
 
-* See :py:mod:`ptbench.data.montgomery` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-    * PNG images 8 bit grayscale
-    * resolution: 4020 x 4892 px or 4892 x 4020 px
-
-* Output image:
-
-    * Transforms:
-
-        * Load raw PNG with :py:mod:`PIL`
-        * Remove black borders
-        * Torch center cropping to get square image
-
-    * Final specifications
-
-        * Grayscale (single channel), 8 bits
-        * Varying resolutions
+See :py:mod:`ptbench.data.montgomery.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/montgomery/fold_1.py b/src/ptbench/data/montgomery/fold_1.py
index 0a74516f..3698a9ed 100644
--- a/src/ptbench/data/montgomery/fold_1.py
+++ b/src/ptbench/data/montgomery/fold_1.py
@@ -4,27 +4,7 @@
 
 """Montgomery datamodule for TB detection (default protocol)
 
-* See :py:mod:`ptbench.data.montgomery` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-    * PNG images 8 bit grayscale
-    * resolution: 4020 x 4892 px or 4892 x 4020 px
-
-* Output image:
-
-    * Transforms:
-
-        * Load raw PNG with :py:mod:`PIL`
-        * Remove black borders
-        * Torch center cropping to get square image
-
-    * Final specifications
-
-        * Grayscale (single channel), 8 bits
-        * Varying resolutions
+See :py:mod:`ptbench.data.montgomery.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/montgomery/fold_2.py b/src/ptbench/data/montgomery/fold_2.py
index 386d3080..b2d7ac2c 100644
--- a/src/ptbench/data/montgomery/fold_2.py
+++ b/src/ptbench/data/montgomery/fold_2.py
@@ -4,27 +4,7 @@
 
 """Montgomery datamodule for TB detection (default protocol)
 
-* See :py:mod:`ptbench.data.montgomery` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-    * PNG images 8 bit grayscale
-    * resolution: 4020 x 4892 px or 4892 x 4020 px
-
-* Output image:
-
-    * Transforms:
-
-        * Load raw PNG with :py:mod:`PIL`
-        * Remove black borders
-        * Torch center cropping to get square image
-
-    * Final specifications
-
-        * Grayscale (single channel), 8 bits
-        * Varying resolutions
+See :py:mod:`ptbench.data.montgomery.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/montgomery/fold_3.py b/src/ptbench/data/montgomery/fold_3.py
index 1bfaa888..1c566e4f 100644
--- a/src/ptbench/data/montgomery/fold_3.py
+++ b/src/ptbench/data/montgomery/fold_3.py
@@ -4,27 +4,7 @@
 
 """Montgomery datamodule for TB detection (default protocol)
 
-* See :py:mod:`ptbench.data.montgomery` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-    * PNG images 8 bit grayscale
-    * resolution: 4020 x 4892 px or 4892 x 4020 px
-
-* Output image:
-
-    * Transforms:
-
-        * Load raw PNG with :py:mod:`PIL`
-        * Remove black borders
-        * Torch center cropping to get square image
-
-    * Final specifications
-
-        * Grayscale (single channel), 8 bits
-        * Varying resolutions
+See :py:mod:`ptbench.data.montgomery.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/montgomery/fold_4.py b/src/ptbench/data/montgomery/fold_4.py
index b955cb15..4b68bd53 100644
--- a/src/ptbench/data/montgomery/fold_4.py
+++ b/src/ptbench/data/montgomery/fold_4.py
@@ -4,27 +4,7 @@
 
 """Montgomery datamodule for TB detection (default protocol)
 
-* See :py:mod:`ptbench.data.montgomery` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-    * PNG images 8 bit grayscale
-    * resolution: 4020 x 4892 px or 4892 x 4020 px
-
-* Output image:
-
-    * Transforms:
-
-        * Load raw PNG with :py:mod:`PIL`
-        * Remove black borders
-        * Torch center cropping to get square image
-
-    * Final specifications
-
-        * Grayscale (single channel), 8 bits
-        * Varying resolutions
+See :py:mod:`ptbench.data.montgomery.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/montgomery/fold_5.py b/src/ptbench/data/montgomery/fold_5.py
index 5604cfdd..59891e8e 100644
--- a/src/ptbench/data/montgomery/fold_5.py
+++ b/src/ptbench/data/montgomery/fold_5.py
@@ -4,27 +4,7 @@
 
 """Montgomery datamodule for TB detection (default protocol)
 
-* See :py:mod:`ptbench.data.montgomery` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-    * PNG images 8 bit grayscale
-    * resolution: 4020 x 4892 px or 4892 x 4020 px
-
-* Output image:
-
-    * Transforms:
-
-        * Load raw PNG with :py:mod:`PIL`
-        * Remove black borders
-        * Torch center cropping to get square image
-
-    * Final specifications
-
-        * Grayscale (single channel), 8 bits
-        * Varying resolutions
+See :py:mod:`ptbench.data.montgomery.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/montgomery/fold_6.py b/src/ptbench/data/montgomery/fold_6.py
index 72178209..e6c1d31a 100644
--- a/src/ptbench/data/montgomery/fold_6.py
+++ b/src/ptbench/data/montgomery/fold_6.py
@@ -4,27 +4,7 @@
 
 """Montgomery datamodule for TB detection (default protocol)
 
-* See :py:mod:`ptbench.data.montgomery` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-    * PNG images 8 bit grayscale
-    * resolution: 4020 x 4892 px or 4892 x 4020 px
-
-* Output image:
-
-    * Transforms:
-
-        * Load raw PNG with :py:mod:`PIL`
-        * Remove black borders
-        * Torch center cropping to get square image
-
-    * Final specifications
-
-        * Grayscale (single channel), 8 bits
-        * Varying resolutions
+See :py:mod:`ptbench.data.montgomery.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/montgomery/fold_7.py b/src/ptbench/data/montgomery/fold_7.py
index de895133..44dd8051 100644
--- a/src/ptbench/data/montgomery/fold_7.py
+++ b/src/ptbench/data/montgomery/fold_7.py
@@ -4,27 +4,7 @@
 
 """Montgomery datamodule for TB detection (default protocol)
 
-* See :py:mod:`ptbench.data.montgomery` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-    * PNG images 8 bit grayscale
-    * resolution: 4020 x 4892 px or 4892 x 4020 px
-
-* Output image:
-
-    * Transforms:
-
-        * Load raw PNG with :py:mod:`PIL`
-        * Remove black borders
-        * Torch center cropping to get square image
-
-    * Final specifications
-
-        * Grayscale (single channel), 8 bits
-        * Varying resolutions
+See :py:mod:`ptbench.data.montgomery.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/montgomery/fold_8.py b/src/ptbench/data/montgomery/fold_8.py
index bc901fc5..fd7edde6 100644
--- a/src/ptbench/data/montgomery/fold_8.py
+++ b/src/ptbench/data/montgomery/fold_8.py
@@ -4,27 +4,7 @@
 
 """Montgomery datamodule for TB detection (default protocol)
 
-* See :py:mod:`ptbench.data.montgomery` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-    * PNG images 8 bit grayscale
-    * resolution: 4020 x 4892 px or 4892 x 4020 px
-
-* Output image:
-
-    * Transforms:
-
-        * Load raw PNG with :py:mod:`PIL`
-        * Remove black borders
-        * Torch center cropping to get square image
-
-    * Final specifications
-
-        * Grayscale (single channel), 8 bits
-        * Varying resolutions
+See :py:mod:`ptbench.data.montgomery.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/montgomery/fold_9.py b/src/ptbench/data/montgomery/fold_9.py
index 758ae340..91228362 100644
--- a/src/ptbench/data/montgomery/fold_9.py
+++ b/src/ptbench/data/montgomery/fold_9.py
@@ -4,27 +4,7 @@
 
 """Montgomery datamodule for TB detection (default protocol)
 
-* See :py:mod:`ptbench.data.montgomery` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-    * PNG images 8 bit grayscale
-    * resolution: 4020 x 4892 px or 4892 x 4020 px
-
-* Output image:
-
-    * Transforms:
-
-        * Load raw PNG with :py:mod:`PIL`
-        * Remove black borders
-        * Torch center cropping to get square image
-
-    * Final specifications
-
-        * Grayscale (single channel), 8 bits
-        * Varying resolutions
+See :py:mod:`ptbench.data.montgomery.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/shenzhen/default.py b/src/ptbench/data/shenzhen/default.py
index ba8a2b57..a163b9bc 100644
--- a/src/ptbench/data/shenzhen/default.py
+++ b/src/ptbench/data/shenzhen/default.py
@@ -2,33 +2,42 @@
 #
 # SPDX-License-Identifier: GPL-3.0-or-later
 
-"""Shenzhen datamodule for computer-aided diagnosis (default protocol)
+"""Shenzhen datamodule for computer-aided diagnosis (``default`` protocol)
 
-See :py:mod:`ptbench.data.shenzhen` for more database details.
+The standard digital image database for Tuberculosis was created by the National
+Library of Medicine, Maryland, USA in collaboration with Shenzhen No.3 People’s
+Hospital, Guangdong Medical College, Shenzhen, China. The Chest X-rays are from
+out-patient clinics, and were captured as part of the daily routine using
+Philips DR Digital Diagnose systems.
 
-This configuration:
+* Database reference: [MONTGOMERY-SHENZHEN-2014]_
+* Original resolution (height x width or width x height): 3000 x 3000 or less
+* This split:
 
-* Raw data input (on disk):
+  * Split reference: None
+  * Training samples: 64% of TB and healthy CXR (including labels)
+  * Validation samples: 16% of TB and healthy CXR (including labels)
+  * Test samples: 20% of TB and healthy CXR (including labels)
+
+Data specifications:
 
-  * PNG images (grayscale, encoded as RGB images)
-  * Variable width and height:
+* Raw data input (on disk):
 
-    * widths: from 1130 to 3001 pixels
-    * heights: from 948 to 3001 pixels
+    * PNG images (grayscale, encoded as RGB images with "inverted" grayscale scale)
+    * Variable width and height
 
 * Output image:
 
-  * Transforms:
+    * Transforms:
 
-    * Load raw PNG with :py:mod:`PIL`
-    * Remove black borders
-    * Torch center cropping to get square image
+        * Load raw PNG with :py:mod:`PIL`
+        * Remove black borders
+        * Torch center cropping to get square image
 
-  * Final specifications:
+    * Final specifications:
 
-    * grayscale, encoded as RGB image
-    * varying resolutions
-    
+        * Grayscale, encoded as a single plane image, 8 bits
+        * Square, with varying resolutions, depending on the input image
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/shenzhen/fold_0.py b/src/ptbench/data/shenzhen/fold_0.py
index d5c3e447..b5059744 100644
--- a/src/ptbench/data/shenzhen/fold_0.py
+++ b/src/ptbench/data/shenzhen/fold_0.py
@@ -4,31 +4,7 @@
 
 """Shenzhen datamodule for computer-aided diagnosis (fold 0)
 
-See :py:mod:`ptbench.data.shenzhen` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-  * PNG images (grayscale, encoded as RGB images)
-  * Variable width and height:
-
-    * widths: from 1130 to 3001 pixels
-    * heights: from 948 to 3001 pixels
-
-* Output image:
-
-  * Transforms:
-
-    * Load raw PNG with :py:mod:`PIL`
-    * Remove black borders
-    * Torch center cropping to get square image
-
-  * Final specifications:
-
-    * grayscale, encoded as RGB image
-    * varying resolutions
-    
+See :py:mod:`ptbench.data.shenzhen.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/shenzhen/fold_1.py b/src/ptbench/data/shenzhen/fold_1.py
index 5c7a294a..1041c3e4 100644
--- a/src/ptbench/data/shenzhen/fold_1.py
+++ b/src/ptbench/data/shenzhen/fold_1.py
@@ -4,31 +4,7 @@
 
 """Shenzhen datamodule for computer-aided diagnosis (fold 1)
 
-See :py:mod:`ptbench.data.shenzhen` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-  * PNG images (grayscale, encoded as RGB images)
-  * Variable width and height:
-
-    * widths: from 1130 to 3001 pixels
-    * heights: from 948 to 3001 pixels
-
-* Output image:
-
-  * Transforms:
-
-    * Load raw PNG with :py:mod:`PIL`
-    * Remove black borders
-    * Torch center cropping to get square image
-
-  * Final specifications:
-
-    * grayscale, encoded as RGB image
-    * varying resolutions
-    
+See :py:mod:`ptbench.data.shenzhen.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/shenzhen/fold_2.py b/src/ptbench/data/shenzhen/fold_2.py
index 31480af5..5026116a 100644
--- a/src/ptbench/data/shenzhen/fold_2.py
+++ b/src/ptbench/data/shenzhen/fold_2.py
@@ -4,31 +4,7 @@
 
 """Shenzhen datamodule for computer-aided diagnosis (fold 2)
 
-See :py:mod:`ptbench.data.shenzhen` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-  * PNG images (grayscale, encoded as RGB images)
-  * Variable width and height:
-
-    * widths: from 1130 to 3001 pixels
-    * heights: from 948 to 3001 pixels
-
-* Output image:
-
-  * Transforms:
-
-    * Load raw PNG with :py:mod:`PIL`
-    * Remove black borders
-    * Torch center cropping to get square image
-
-  * Final specifications:
-
-    * grayscale, encoded as RGB image
-    * varying resolutions
-    
+See :py:mod:`ptbench.data.shenzhen.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/shenzhen/fold_3.py b/src/ptbench/data/shenzhen/fold_3.py
index a1881ad6..16c00157 100644
--- a/src/ptbench/data/shenzhen/fold_3.py
+++ b/src/ptbench/data/shenzhen/fold_3.py
@@ -4,31 +4,7 @@
 
 """Shenzhen datamodule for computer-aided diagnosis (fold 3)
 
-See :py:mod:`ptbench.data.shenzhen` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-  * PNG images (grayscale, encoded as RGB images)
-  * Variable width and height:
-
-    * widths: from 1130 to 3001 pixels
-    * heights: from 948 to 3001 pixels
-
-* Output image:
-
-  * Transforms:
-
-    * Load raw PNG with :py:mod:`PIL`
-    * Remove black borders
-    * Torch center cropping to get square image
-
-  * Final specifications:
-
-    * grayscale, encoded as RGB image
-    * varying resolutions
-    
+See :py:mod:`ptbench.data.shenzhen.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/shenzhen/fold_4.py b/src/ptbench/data/shenzhen/fold_4.py
index f86eb665..c0b0fdac 100644
--- a/src/ptbench/data/shenzhen/fold_4.py
+++ b/src/ptbench/data/shenzhen/fold_4.py
@@ -4,31 +4,7 @@
 
 """Shenzhen datamodule for computer-aided diagnosis (fold 4)
 
-See :py:mod:`ptbench.data.shenzhen` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-  * PNG images (grayscale, encoded as RGB images)
-  * Variable width and height:
-
-    * widths: from 1130 to 3001 pixels
-    * heights: from 948 to 3001 pixels
-
-* Output image:
-
-  * Transforms:
-
-    * Load raw PNG with :py:mod:`PIL`
-    * Remove black borders
-    * Torch center cropping to get square image
-
-  * Final specifications:
-
-    * grayscale, encoded as RGB image
-    * varying resolutions
-    
+See :py:mod:`ptbench.data.shenzhen.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/shenzhen/fold_5.py b/src/ptbench/data/shenzhen/fold_5.py
index 16ea44f7..0397955e 100644
--- a/src/ptbench/data/shenzhen/fold_5.py
+++ b/src/ptbench/data/shenzhen/fold_5.py
@@ -4,31 +4,7 @@
 
 """Shenzhen datamodule for computer-aided diagnosis (fold 5)
 
-See :py:mod:`ptbench.data.shenzhen` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-  * PNG images (grayscale, encoded as RGB images)
-  * Variable width and height:
-
-    * widths: from 1130 to 3001 pixels
-    * heights: from 948 to 3001 pixels
-
-* Output image:
-
-  * Transforms:
-
-    * Load raw PNG with :py:mod:`PIL`
-    * Remove black borders
-    * Torch center cropping to get square image
-
-  * Final specifications:
-
-    * grayscale, encoded as RGB image
-    * varying resolutions
-    
+See :py:mod:`ptbench.data.shenzhen.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/shenzhen/fold_6.py b/src/ptbench/data/shenzhen/fold_6.py
index caecaa0d..145685ea 100644
--- a/src/ptbench/data/shenzhen/fold_6.py
+++ b/src/ptbench/data/shenzhen/fold_6.py
@@ -4,31 +4,7 @@
 
 """Shenzhen datamodule for computer-aided diagnosis (fold 6)
 
-See :py:mod:`ptbench.data.shenzhen` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-  * PNG images (grayscale, encoded as RGB images)
-  * Variable width and height:
-
-    * widths: from 1130 to 3001 pixels
-    * heights: from 948 to 3001 pixels
-
-* Output image:
-
-  * Transforms:
-
-    * Load raw PNG with :py:mod:`PIL`
-    * Remove black borders
-    * Torch center cropping to get square image
-
-  * Final specifications:
-
-    * grayscale, encoded as RGB image
-    * varying resolutions
-    
+See :py:mod:`ptbench.data.shenzhen.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/shenzhen/fold_7.py b/src/ptbench/data/shenzhen/fold_7.py
index ee34486c..5b8d7403 100644
--- a/src/ptbench/data/shenzhen/fold_7.py
+++ b/src/ptbench/data/shenzhen/fold_7.py
@@ -4,31 +4,7 @@
 
 """Shenzhen datamodule for computer-aided diagnosis (fold 7)
 
-See :py:mod:`ptbench.data.shenzhen` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-  * PNG images (grayscale, encoded as RGB images)
-  * Variable width and height:
-
-    * widths: from 1130 to 3001 pixels
-    * heights: from 948 to 3001 pixels
-
-* Output image:
-
-  * Transforms:
-
-    * Load raw PNG with :py:mod:`PIL`
-    * Remove black borders
-    * Torch center cropping to get square image
-
-  * Final specifications:
-
-    * grayscale, encoded as RGB image
-    * varying resolutions
-    
+See :py:mod:`ptbench.data.shenzhen.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/shenzhen/fold_8.py b/src/ptbench/data/shenzhen/fold_8.py
index 6c81ca70..e9ce1a2f 100644
--- a/src/ptbench/data/shenzhen/fold_8.py
+++ b/src/ptbench/data/shenzhen/fold_8.py
@@ -4,31 +4,7 @@
 
 """Shenzhen datamodule for computer-aided diagnosis (fold 8)
 
-See :py:mod:`ptbench.data.shenzhen` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-  * PNG images (grayscale, encoded as RGB images)
-  * Variable width and height:
-
-    * widths: from 1130 to 3001 pixels
-    * heights: from 948 to 3001 pixels
-
-* Output image:
-
-  * Transforms:
-
-    * Load raw PNG with :py:mod:`PIL`
-    * Remove black borders
-    * Torch center cropping to get square image
-
-  * Final specifications:
-
-    * grayscale, encoded as RGB image
-    * varying resolutions
-    
+See :py:mod:`ptbench.data.shenzhen.default` for input/output details.
 """
 
 import importlib.resources
diff --git a/src/ptbench/data/shenzhen/fold_9.py b/src/ptbench/data/shenzhen/fold_9.py
index 546e449e..6da8dd3d 100644
--- a/src/ptbench/data/shenzhen/fold_9.py
+++ b/src/ptbench/data/shenzhen/fold_9.py
@@ -4,31 +4,7 @@
 
 """Shenzhen datamodule for computer-aided diagnosis (fold 9)
 
-See :py:mod:`ptbench.data.shenzhen` for more database details.
-
-This configuration:
-
-* Raw data input (on disk):
-
-  * PNG images (grayscale, encoded as RGB images)
-  * Variable width and height:
-
-    * widths: from 1130 to 3001 pixels
-    * heights: from 948 to 3001 pixels
-
-* Output image:
-
-  * Transforms:
-
-    * Load raw PNG with :py:mod:`PIL`
-    * Remove black borders
-    * Torch center cropping to get square image
-
-  * Final specifications:
-
-    * grayscale, encoded as RGB image
-    * varying resolutions
-    
+See :py:mod:`ptbench.data.shenzhen.default` for input/output details.
 """
 
 import importlib.resources
-- 
GitLab