Skip to content

Commit 04a3288

Browse files
kostrykinkostrykin
authored
Merge ISBI'24 implementation, examples, and documentation (#2)
* adds `fast2` mode for `global-energy-minimization` * Update README.rst Add ISBI'24 publication reference * Add BBBC039 dataset * Clean up `global-energy-minimization` stage Replace `global-energy-minimization/mode` hyperparameter by `global-energy-minimization/pruning` * Add ISBI'24 publication reference to documentation * Add examples for BBBC039 dataset * Increment version to 0.2.0 * Fix versions of dependencies * Update README.rst * Update LICENSE * Update README.rst * Update README.rst * Update README.rst * Update README.rst * Update README.rst * Update README.rst * Update README.rst * Move RTD requirements to dedicated file * Update README.rst * Update README.rst * Update README.rst * Add missing examples for ISBI'24 paper * Fix .readthedocs.yaml See: https://stackoverflow.com/a/65006808/1444073 * Fix dependencies for RTD * Update pipeline.rst * Update dsmcfg.py * Update pipeline.rst * Update dsmcfg.py * Update globalenergymin.py * Update c2freganal.py * Update minsetcover.py * Update objects.py * Update automation.py * Update dsm.py * Update postprocess.py * Update pipeline.py
1 parent d5c1d3f commit 04a3288

File tree

26 files changed

+431
-97
lines changed

26 files changed

+431
-97
lines changed

‎.readthedocs.yaml

+1
Original file line numberDiff line numberDiff line change
@@ -12,5 +12,6 @@ python:
1212
# Install our python package before building the docs
1313
install:
1414
- requirements: requirements.txt
15+
- requirements: docs/requirements.txt
1516
- method: pip
1617
path: .

‎LICENSE

+1-1
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@ MIT License
22

33
This license applies to the entire repository except for sub-directories which have their own LICENSE file.
44

5-
Copyright (c) 2017-2023 Leonid Kostrykin, Biomedical Computer Vision Group, Heidelberg University
5+
Copyright (c) 2017-2024 Leonid Kostrykin, Biomedical Computer Vision Group, Heidelberg University
66

77
Permission is hereby granted, free of charge, to any person obtaining a copy
88
of this software and associated documentation files (the "Software"), to deal

‎README.rst

+25-6
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
1-
SuperDSM
2-
========
1+
`SuperDSM <https://github.com/BMCV/SuperDSM>`_
2+
==============================================
33

44
.. image:: https://img.shields.io/badge/Install%20with-conda-%2387c305
55
:target: https://anaconda.org/bioconda/superdsm
@@ -22,16 +22,35 @@ The documentation is available here: https://superdsm.readthedocs.io
2222

2323
Use ``python -m unittest`` in the root directory of the repository to run the test suite.
2424

25-
For full performance on both Intel and AMD CPUs, NumPy with MKL support must be used (instead of OpenBLAS which is often the default). When using the source tree instead of the Conda package from Bioconda, this can be ensured by adding the dependency ``blas =*=mkl`` to your Conda environment.
25+
Dependency Version Considerations:
26+
""""""""""""""""""""""""""""""""""
27+
28+
The file *superdsm.yml* specifies the Conda environment required to accurately reproduce the results from our publications. For most of the dependencies (maybe even all), newer versions are also known to work, however, it has been observed that using different versions might yield slightly different results. To enhance consistency, reproducibility, and `FAIRness <https://www.nature.com/articles/s41597-022-01710-x>`_, most dependency versions are thus pinned.
29+
30+
This also concerns BLAS, which is pinned to ``blas==1.0``. As an alternative to using Conda, *requirements.txt* specifies the required *pip* dependencies with pinned versions. However, to the best of our knowledge, it is not possible to request a specific BLAS version in *pip*, meaning that using *pip* instead of Conda is discouraged.
31+
32+
Note that our Conda package from Bioconda allows different dependency versions, because otherwise it would not be possible to use the package with newer versions of Python. Thus, when using our Conda package, keep in mind that sticking to the versions of the dependencies specified in *superdsm.yml* is recommended.
33+
34+
Performance Considerations:
35+
"""""""""""""""""""""""""""
36+
37+
For full performance on both Intel and AMD CPUs, NumPy with MKL support must be used (instead of OpenBLAS which is often the default, see `details <https://stackoverflow.com/questions/62783262/why-is-numpy-with-ryzen-threadripper-so-much-slower-than-xeon>`_). When using Conda, this can be ensured by adding the dependency ``blas=1.0=mkl`` to the Conda environment (or ``blas=*=mkl`` to allow different versions).
38+
39+
To take advantage of the acceleration provided by MKL on AMD CPUs, the MKL version must be pinned to ``2020.0``. Both specifications are included in the Conda environment specified in *superdsm.yml*. In addition, the environment variable ``MKL_DEBUG_CPU_TYPE=5`` must be set.
40+
41+
Later versions of MKL do not support ``MKL_DEBUG_CPU_TYPE=5``, and previous versions do not offer the required APIs. Unfortunately, it looks like this particular version of MKL has been removed from PyPI (see `available versions <https://pypi.org/project/mkl/#history>`_), so it is not possible to gain the full performance on AMD CPUs using *pip* instead of Conda, and thus the version of MKL is not pinned in *requirements.txt* by default.
2642

2743
Publications:
44+
"""""""""""""
45+
46+
* L\. Kostrykin and K\. Rohr, *"Robust Graph Pruning for Efficient Segmentation and Cluster Splitting of Cell Nuclei using Deformable Shape Models,"* accepted for presentation at *IEEE International Symposium on Biomedical Imaging (ISBI)*, Athens, Greece, May 27–30, 2024.
2847

29-
* `L. Kostrykin and K. Rohr, "Superadditivity and Convex Optimization for Globally Optimal Cell Segmentation Using Deformable Shape Models," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 45(3), pp. 3831–3847, 2023.
30-
<https://doi.org/10.1109/TPAMI.2022.3185583>`_
48+
* L\. Kostrykin and K\. Rohr, *"Superadditivity and Convex Optimization for Globally Optimal Cell Segmentation Using Deformable Shape Models,"* in *IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)*, vol. 45(3), pp. 3831–3847, 2023.
49+
`[doi] <https://doi.org/10.1109/TPAMI.2022.3185583>`_
3150

3251
----
3352

34-
Copyright (c) 2017-2023 Leonid Kostrykin, Biomedical Computer Vision Group, Heidelberg University
53+
Copyright (c) 2017-2024 Leonid Kostrykin, Biomedical Computer Vision Group, Heidelberg University
3554

3655
This work is licensed under the terms of the MIT license.
3756
For a copy, see `LICENSE </LICENSE>`_.

‎docs/requirements.txt

+2
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,2 @@
1+
sphinx-autorun>=1.1.1
2+
sphinx-rtd-theme

‎docs/source/pipeline.rst

+1-1
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@ Refer to the :py:mod:`.pipeline` module for a general overview of the pipeline c
1010
Theory
1111
------
1212

13-
This is an overview of the fundamental concepts described in the paper (:ref:`Kostrykin and Rohr, 2023 <references>`).
13+
This is an overview of the fundamental concepts described in :ref:`Kostrykin and Rohr (TPAMI 2023) <references>`.
1414

1515
.. _pipeline_theory_dsm:
1616

‎docs/source/usage.rst

+4-2
Original file line numberDiff line numberDiff line change
@@ -75,5 +75,7 @@ References
7575

7676
If you use SuperDSM, please cite:
7777

78-
`L. Kostrykin and K. Rohr, "Superadditivity and Convex Optimization for Globally Optimal Cell Segmentation Using Deformable Shape Models," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 45(3), pp. 3831–3847, 2023.
79-
<https://doi.org/10.1109/TPAMI.2022.3185583>`_
78+
* L\. Kostrykin and K\. Rohr, *"Robust Graph Pruning for Efficient Segmentation and Cluster Splitting of Cell Nuclei using Deformable Shape Models,"* accepted for presentation at *IEEE International Symposium on Biomedical Imaging (ISBI)*, Athens, Greece, May 27–30, 2024.
79+
80+
* L\. Kostrykin and K\. Rohr, *"Superadditivity and Convex Optimization for Globally Optimal Cell Segmentation Using Deformable Shape Models,"* in *IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)*, vol. 45(3), pp. 3831–3847, 2023.
81+
`[doi] <https://doi.org/10.1109/TPAMI.2022.3185583>`_

‎examples/BBBC039/isbi24/task.json

+9
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,9 @@
1+
{
2+
"runnable": true,
3+
4+
"config": {
5+
"global-energy-minimization": {
6+
"pruning": "isbi24"
7+
}
8+
}
9+
}

‎examples/BBBC039/task.json

+224
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,224 @@
1+
{
2+
"runnable": true,
3+
"num_cpus": 16,
4+
"environ": {
5+
"MKL_NUM_THREADS": 2,
6+
"OPENBLAS_NUM_THREADS": 2
7+
},
8+
9+
"img_pathpattern": "{ROOTDIR}/../data/{DIRNAME}/IXMtest_%s.tif",
10+
"seg_pathpattern": "seg/%s.png",
11+
"adj_pathpattern": "adj/%s.png",
12+
"log_pathpattern": "log/%s",
13+
"cfg_pathpattern": "cfg/%s.json",
14+
"overlay_pathpattern": "overlays/%s.png",
15+
"file_ids": [
16+
"A02_s1_w1051DAA7C-7042-435F-99F0-1E847D9B42CB",
17+
"A06_s6_w1B9577918-4973-4A87-BA73-A168AA755527",
18+
"A09_s1_w1CE70AD49-290D-4312-82E6-CDC717F32637",
19+
"A12_s7_w1EAEEA614-51ED-43B3-A4FF-088730911E4C",
20+
"A15_s5_w1825174D4-ED30-490C-9635-6196417D6C9D",
21+
"A16_s2_w15AF20A10-82AE-48FA-AC50-7AE8AC3AA544",
22+
"A16_s3_w1032BE329-E21B-4E1B-B4B8-58700685EE0C",
23+
"A18_s1_w1BFDF1C94-9C1F-4F5F-BBC1-05196333B1BF",
24+
"A20_s4_w153DE191F-B112-471B-AC0A-FAD27990714B",
25+
"A21_s1_w1D8BF2790-C628-4109-94AF-DF6D8A3480B6",
26+
"A22_s8_w1E2AFE190-831D-4D9C-961E-3AA2ECB3599D",
27+
"A24_s9_w152CD6793-DA45-4975-869A-CBD49D645E37",
28+
"B02_s9_w124B5080D-EBE1-47D2-B147-C0F342039EDF",
29+
"B04_s2_w17C6C7F8D-98F7-422B-92CD-EA61EE813325",
30+
"B04_s4_w1F6AEFA0F-AF87-4B3B-A334-698647CFE043",
31+
"B05_s5_w10296BEF4-0F48-4EC8-BC24-6C604871398B",
32+
"B12_s2_w19F7E0279-D087-4B5E-9899-61971C29CB78",
33+
"B12_s9_w1ABE4C6B8-6110-4D7D-B782-94C43FDC2E1F",
34+
"B17_s7_w1215A0A98-4A76-4846-B54A-F7C1EAF84E02",
35+
"B19_s7_w1E43B84DB-39E2-4BFB-8CB4-554B32098C75",
36+
"B20_s2_w159B9FE71-035A-4DED-B0CA-C76916C968BC",
37+
"B21_s3_w141E785B1-44FE-4ED0-9CCE-6FF076EFE9FE",
38+
"B21_s4_w1521471E0-9BD7-492A-8739-9C782C2585B0",
39+
"B21_s7_w1276972A2-1794-4A98-8E76-876A597C346B",
40+
"B22_s6_w19397283D-2426-4942-8A0F-EB75E1D48C2D",
41+
"B22_s8_w10754C18F-B059-47B4-A423-FF429B984D80",
42+
"B23_s1_w152C21D3B-75A9-4FF9-839A-23E6CE19AEEC",
43+
"B24_s9_w18C4FE0DD-12CA-4711-9722-3E3105D1E691",
44+
"C02_s4_w1661FF37C-AB9B-4038-BDF3-D21DFD2811BE",
45+
"C05_s7_w1F71963FB-8F29-41CB-A5F5-07CB9584BBC5",
46+
"C07_s5_w12C9F2926-A017-4962-8660-72C9C20C86E4",
47+
"C09_s7_w1768A3B0D-47FE-4D77-B1C6-46018E29486F",
48+
"C14_s8_w1612E3D43-FA7E-4FFB-BE53-F34A3A0CAFCD",
49+
"C18_s1_w11C16FC59-2E29-496A-803A-89581FDF538A",
50+
"C19_s4_w1AD3DC23E-B9B5-46E9-9716-26096C672A13",
51+
"C21_s9_w1289444BE-8623-4867-86E7-01E155C6538D",
52+
"C23_s2_w10533E965-3DC3-4556-9C26-EA8DD4F8DE61",
53+
"D01_s1_w181AE787C-B5EE-4150-A90E-2FE43165C32E",
54+
"D02_s8_w1AC6783DF-ED35-4818-8091-E6D02AF4BFBD",
55+
"D04_s2_w1E05C134E-BAFE-4FD7-8116-2D05E0839879",
56+
"D04_s9_w17B6268DB-8215-4DC6-943C-CC009A8A5732",
57+
"D06_s5_w13C67AAA9-6E81-42DB-AC5F-7126602F3607",
58+
"D06_s8_w1BE84C8EF-4CD0-4B56-8267-4E662F57AC25",
59+
"D07_s4_w16CF58D03-0B05-41FE-AE73-7298887DEBB1",
60+
"D08_s7_w1F476A544-A07C-4E9B-A6CB-BAEF4CD6F64E",
61+
"D10_s2_w1BB72A093-72AB-476C-9513-2FF43BFB2134",
62+
"D19_s6_w1EB1F11AE-4FB6-481F-94D9-40246870F0CB",
63+
"D20_s3_w19D371AF3-0189-48A3-AF3B-D108DE6A017F",
64+
"D20_s9_w1588B4C42-5A0D-4B93-9A80-A6527CC1C411",
65+
"E04_s4_w1F5ABC12B-FABD-4E33-AB99-82271209853E",
66+
"E05_s2_w15CCE97F8-3F02-412E-8BF2-FC92972DDA1A",
67+
"E06_s3_w1701573EB-CE9A-4D76-8668-0416996E1DCD",
68+
"E07_s1_w1641C6847-142B-463F-8B08-5B3296615572",
69+
"E08_s2_w1588EDEBA-1BBA-4E49-810F-2EF2064C8846",
70+
"E12_s9_w1A811DEC0-ADD9-411A-B5D5-A654C70F253D",
71+
"E15_s5_w1419F04CC-50C3-41FB-A87A-FB17647A2E33",
72+
"E19_s2_w1752F162C-2C91-4A67-BF4F-D073C3D88291",
73+
"E19_s4_w129545707-4CD3-4498-AC27-E4AE24D0253C",
74+
"E23_s8_w10F1F106B-057A-482C-93BB-C9A0B044C054",
75+
"F01_s2_w189F7DC13-4572-43C9-8C23-CD9AEB41B24A",
76+
"F03_s5_w1FE22EC7D-720A-4848-888A-6AFE1E1C85B0",
77+
"F04_s5_w1D94DA1A2-873C-44B3-80EB-36DC2A97E9A3",
78+
"F07_s5_w15678F19D-6F23-4FB4-8851-5B57CF8CCCF8",
79+
"F08_s1_w144C3056F-C4DD-4D39-A40F-4F4576A6DBD8",
80+
"F08_s7_w117D78658-1BA7-495D-9EAA-E4FA155861AC",
81+
"F12_s5_w17F3E9DFC-6705-40A9-B5FE-C60261D73052",
82+
"F12_s8_w1DBD80811-5297-4415-ACD1-EC9286BE76A4",
83+
"F12_s9_w1CD127D58-4641-41D3-88CC-E12B7FE77D78",
84+
"F14_s7_w1E67F2A8A-EEC1-4F01-A1E2-0ECFC639637F",
85+
"F21_s1_w1E2770CF6-417B-4804-9B45-6EE7D783D8CA",
86+
"F22_s6_w1D6C33377-3947-4DAC-B62D-531C2565F150",
87+
"F22_s6_w1F4C7ADE4-B68D-4D30-A063-722B87AA2DA1",
88+
"G01_s2_w1FBE52723-8BDF-4346-89BB-216A4A69ED1C",
89+
"G05_s6_w1AA0013DE-ED7A-428E-A17A-09959E7DC1A2",
90+
"G06_s3_w19ABE30DE-4BFC-4CD5-AD16-6C8BD6E69F9D",
91+
"G06_s7_w19444140C-EF22-42DE-863A-514D836BE850",
92+
"G08_s6_w1823194BE-DE9A-4638-BACD-CABE5FA4C89C",
93+
"G10_s3_w1C1257E17-1DBA-4619-B06E-D6DBB8A53088",
94+
"G12_s6_w16850371E-A405-4D73-9816-F5F68F885D38",
95+
"G13_s9_w19606195E-12B4-46FF-9B83-1F7FE11B3AFB",
96+
"G16_s3_w1B4690283-D75E-4DFB-92B0-29419E8292C6",
97+
"G16_s6_w1C3F175E2-0AF5-401C-AC5C-8E128A237B22",
98+
"G18_s9_w17F495AF2-03D2-4683-BD37-56887F4A3A84",
99+
"G22_s3_w157F2847B-C953-410E-8F60-956A6023AED4",
100+
"G22_s5_w1CE8AEFCD-7739-4D60-B112-9D2D73EE05E5",
101+
"G23_s4_w126F855D5-89CE-432B-9592-684B6E5EEF24",
102+
"G23_s8_w12910C37A-C52C-45B2-A796-7C60A2247C32",
103+
"H01_s8_w13C7E7AAD-FC32-42F8-88C5-DDEA4BD8832C",
104+
"H01_s9_w1237AA324-894F-4C97-99F3-FEF47F52A550",
105+
"H06_s4_w16261E805-AF5F-4785-BB65-C77B61FF39F7",
106+
"H06_s6_w1C3C98946-0210-4D6A-831E-5681452CCD1D",
107+
"H07_s2_w1D8F30687-16A2-4A0D-BDD5-377257089055",
108+
"H08_s8_w1418B93F8-515E-49A7-A9AE-CAF7A1589A38",
109+
"H11_s5_w1612B789B-6279-41E1-B4C6-009BF5772334",
110+
"H11_s6_w19DF4E879-8DE4-45D6-840B-305BDDB27076",
111+
"H16_s4_w16207B133-B3F5-4C39-87A6-D71359B27581",
112+
"H17_s1_w10A751E6C-5D06-4147-AB73-7FFAE0B57CC5",
113+
"H18_s7_w1D78A9181-AF11-4219-9C1E-2DB2684E948E",
114+
"H24_s6_w15034AA9F-051A-4BF0-A62F-ABAE599C4867",
115+
"I01_s4_w1218CC565-C87E-4390-936A-4D3E51BC10DB",
116+
"I03_s4_w1C724E24D-7AD5-4166-A810-DC75A825495D",
117+
"I04_s9_w16A5CC270-8B92-42EE-AA4A-855776F7D46B",
118+
"I07_s4_w1F156255A-3842-46FB-ABF2-9D041E523F86",
119+
"I08_s2_w11996D679-5D76-4FB8-A681-2014A8999EC8",
120+
"I11_s6_w1B2DC04C7-2D7D-45C6-9DC2-66D8605FBE63",
121+
"I12_s1_w18A06F07F-CD8F-4056-B041-3967BE48AD78",
122+
"I15_s5_w11265571E-AA95-4E58-82CF-A75E3F240956",
123+
"I15_s8_w196F01FA4-CE06-44AE-AAA5-ADEC35BAE513",
124+
"I16_s9_w1582FD22A-5270-4FDC-868F-5F75808E2321",
125+
"I17_s6_w1EB7CA00F-DF95-466C-BF8C-6304B6A4974E",
126+
"I17_s7_w1BB3D534E-DF08-4F21-BAC6-6B105BB40AF9",
127+
"I18_s3_w1544C8B7A-E092-4F9D-B8D3-C489638D770F",
128+
"I23_s5_w1E3053D6B-8CEF-48E3-A6A5-2F0D7C1AA177",
129+
"J02_s5_w1F53DDD35-C0B2-4E39-BA3B-9F66D289AB02",
130+
"J02_s8_w1D9C198F9-ECF0-4EF7-848D-AC7782CD3C28",
131+
"J03_s4_w1472990BF-A442-497B-8A43-8EE065002EEB",
132+
"J07_s9_w16921AB3C-1E9C-41C5-ADDA-7DE5CBF9CCBB",
133+
"J08_s2_w1C146DB1C-05B3-49EF-9C62-1185FD9897AC",
134+
"J11_s2_w1177A125B-DEBE-4CB7-8F26-53A871785BA5",
135+
"J15_s9_w1857B395F-AE7A-4319-BC26-3A041C0E31E9",
136+
"J16_s8_w1AD7BC3EB-6A29-4A53-AA59-E39653839B0D",
137+
"J17_s1_w19E7BCD2F-6D7E-4B39-9154-F93E4D539183",
138+
"J20_s1_w1EEE65E52-7AD8-47C7-A286-6E84C5D77953",
139+
"J20_s5_w15709AA91-01B9-4959-963A-9D1D574D12E4",
140+
"K01_s3_w1054711D8-52CF-4672-B3A0-EC9EC2E681C4",
141+
"K01_s5_w1A3DE001A-72D6-4321-8B25-4300AB0207AC",
142+
"K02_s3_w1FE99F6CA-8D35-4496-8F52-DBE55EFAB7E6",
143+
"K03_s5_w1DC4CE558-042C-482E-8CAE-FCCB57AA9A55",
144+
"K05_s7_w1816D5E2A-B9D8-48DA-9FE6-CBC35FE38C8D",
145+
"K05_s9_w1647AEF60-D7B0-4D83-AFA5-2B05E968468F",
146+
"K06_s9_w1787DFDA5-E2CE-40DB-9430-CC3766599BCC",
147+
"K11_s4_w139A2D71E-EC27-49CC-BDA0-1118747BDC76",
148+
"K12_s1_w193D6C057-1AA9-4E2F-86EA-2E71961BE68B",
149+
"K12_s6_w160D86D6B-648B-433E-9776-8A42DF40E5FB",
150+
"K12_s7_w12A7857A5-3C92-4A08-8E81-2CA8A99F67AE",
151+
"K17_s5_w1BB3AD9D7-3A85-458C-AB1C-07A816C2C203",
152+
"K18_s8_w11FC8D3EA-E30D-46B8-A45E-0399327AA8AC",
153+
"L01_s3_w1E7E0D198-5FB4-4E10-A27C-C46463DA9E06",
154+
"L03_s2_w1AC4550E2-F824-4A58-9CC5-952AD9ECE76A",
155+
"L03_s6_w1BE79472D-5E2C-422F-A16E-7AC0691C0FD8",
156+
"L05_s2_w1B9C6FAC9-9D48-4184-8D9B-ABFC3BEC1125",
157+
"L06_s4_w14DD3575A-627B-4A71-91F0-3396DE0B33C4",
158+
"L06_s5_w1610AE208-D6E8-4DC4-AC19-D26284AA8556",
159+
"L11_s4_w13C057BB5-9CFB-471F-84B5-72F80654CF81",
160+
"L13_s5_w14B7DEC0F-735B-428F-B8B4-F4FF32B6ED4A",
161+
"L14_s5_w14B42C89E-7650-44AC-9D7B-50BE61EA307E",
162+
"L17_s1_w1DDC627E5-ADF1-441C-A437-D1D91CC0D498",
163+
"L21_s5_w122478CD2-80DC-4B4E-9BC8-A6F6239F4103",
164+
"M04_s3_w1A599DF67-1E7F-4A09-84FD-0080767A735C",
165+
"M12_s5_w16D817EFA-C3C9-45DB-AC15-61BB143DEC62",
166+
"M12_s7_w193E9BA0F-274A-4EE8-A0F0-6A58471A333A",
167+
"M15_s8_w113217199-6161-475C-A786-265D0444C84B",
168+
"M20_s3_w15C73A7C7-F81B-4583-AB8F-0A64336AF070",
169+
"M23_s8_w118BC311D-A998-4161-8256-22839B2421F2",
170+
"N02_s8_w1B14E70A6-CACF-425C-A958-1285482357F3",
171+
"N07_s5_w1D30ED7AB-503E-479D-B5BC-B66472568DE2",
172+
"N08_s2_w159DEEDD5-2882-4F9E-9048-F5BD2B918651",
173+
"N11_s4_w142A84EA3-47C3-4B49-B6CA-BBC6685BBE1E",
174+
"N12_s7_w166EF3FAB-EA33-4B28-91E3-034A1654BAAE",
175+
"N12_s9_w1C98EC3CC-2F8E-4761-A21A-EBF14469AD40",
176+
"N15_s3_w148785C31-5B78-47BA-8802-FB8F2CEA6BE2",
177+
"N15_s8_w16DBEFA45-FE5A-4233-A1B9-89CA8CF81FB8",
178+
"N18_s2_w1CC5ED51D-86C5-437D-8EDD-E56E4C949B3B",
179+
"N18_s5_w1A2FA7E05-3145-461D-9111-FAB41A7F6D6F",
180+
"N21_s1_w14F264B20-62D6-47EC-9BCD-B9416AB302D9",
181+
"N23_s2_w16380EDDA-E5B8-4168-8BE7-F83AD9810680",
182+
"N24_s7_w1AC0733FE-E0FD-45B2-930B-FEEAB052DB36",
183+
"O01_s6_w11A23978B-BAAD-4287-B1F6-FFBCF45C5E2F",
184+
"O02_s4_w1814603DA-DD56-44EF-98C5-AFF3B636F986",
185+
"O04_s2_w19A18B0D8-8E8B-4572-A5C0-5C678F3AD54C",
186+
"O07_s2_w148B3D2F0-D4D6-4F1A-88D3-F18574F52153",
187+
"O07_s6_w17884A281-85EA-4DD8-9D5A-42427988AE36",
188+
"O09_s2_w133C7EDCE-1C7C-41A6-9E52-7AD499E7CDC8",
189+
"O10_s8_w18F4DB020-BFB7-4F13-B99C-C39F8E54F85D",
190+
"O13_s3_w12D9C1C9C-C582-4080-B9BE-4807FA3E0843",
191+
"O15_s6_w1A4844AB3-6D66-4B9A-8212-758E061C349F",
192+
"O16_s2_w1F6F6A3A1-99E4-4029-B734-022806CF6D42",
193+
"O16_s3_w129E61E12-E398-496E-9408-092D2D70E74B",
194+
"O18_s1_w1C2EE9CCF-5055-4E56-BB3A-212A63545A90",
195+
"O18_s7_w19C30A212-58D3-4030-AA4F-B0C4482F1F8A",
196+
"O24_s5_w175443699-7C4C-4E5F-9C5F-57A342CC87DE",
197+
"P01_s3_w1A7DC2612-9C11-4656-B100-102AF8FE8B43",
198+
"P05_s5_w18FC141F8-8BCA-4851-9000-31D080922BDD",
199+
"P07_s8_w144364F25-950A-472C-A529-1A9AD0433B6C",
200+
"P07_s8_w19D682C29-5685-4A33-8CAA-F0797DD7F021",
201+
"P09_s4_w11E382363-8C26-4A73-9D8A-8193D66E5785",
202+
"P10_s2_w130B9EBAE-801C-4AAC-A3CE-1C96EE2E6FF9",
203+
"P10_s7_w1F78192AF-7D11-4D41-80F3-8CD6DB05AB57",
204+
"P13_s6_w1270AEE06-D31F-421C-8147-13644F4C5B94",
205+
"P15_s3_w10F5E9699-743C-4177-93CE-27CFD65A925E",
206+
"P15_s8_w1D26A5BC7-CF59-4027-B785-85AD33773609",
207+
"P19_s6_w10EBAD15B-2822-48B9-B9C8-4DF5D80AB940",
208+
"P21_s4_w19B7DC88B-10BB-4B51-9A10-9E30285688C1",
209+
"P21_s5_w1ACEBEE91-BAFA-49E6-9D97-D07197400A15",
210+
"P23_s7_w13B627CB9-6C57-4049-AAD7-6468A051DD24",
211+
"P23_s9_w1CA355675-EEC7-40B6-82C8-DC7E52A7A825",
212+
"P24_s9_w13AC6C03C-E8D7-4A23-B649-514BB4052F52"
213+
],
214+
215+
"config": {
216+
"AF_scale": 10,
217+
"dsm": {
218+
"AF_alpha": 0.00033
219+
},
220+
"global-energy-minimization": {
221+
"AF_beta": 1.5
222+
}
223+
}
224+
}

‎examples/GOWT1-1/default/adapted/isbi24/task.json

+9
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,9 @@
1+
{
2+
"runnable": true,
3+
4+
"config": {
5+
"global-energy-minimization": {
6+
"pruning": "isbi24"
7+
}
8+
}
9+
}

‎examples/GOWT1-2/default/adapted/isbi24/task.json

+9
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,9 @@
1+
{
2+
"runnable": true,
3+
4+
"config": {
5+
"global-energy-minimization": {
6+
"pruning": "isbi24"
7+
}
8+
}
9+
}

‎examples/NIH3T3/default/adapted/isbi24/task.json

+9
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,9 @@
1+
{
2+
"runnable": true,
3+
4+
"config": {
5+
"global-energy-minimization": {
6+
"pruning": "isbi24"
7+
}
8+
}
9+
}

‎examples/U2OS/default/adapted/isbi24/task.json

+9
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,9 @@
1+
{
2+
"runnable": true,
3+
4+
"config": {
5+
"global-energy-minimization": {
6+
"pruning": "isbi24"
7+
}
8+
}
9+
}

‎examples/load_data.py

+10-3
Original file line numberDiff line numberDiff line change
@@ -26,7 +26,7 @@ def load_data(url, datasets, **kwargs):
2626
src_root = pathlib.Path(archive_dirpath)
2727
for src, dst in datasets:
2828
print(f'Populating {data_dir / dst}')
29-
dst.mkdir(parents=True, exists_ok=True)
29+
pathlib.Path(dst).mkdir(parents=True, exist_ok=True)
3030
shutil.move(str(src_root / src), str(data_dir / dst))
3131

3232
load_data(
@@ -47,13 +47,20 @@ def load_data(url, datasets, **kwargs):
4747
load_data(
4848
'https://bmcv.github.io/SuperDSM/fibroblast-prolif.tar.bz2',
4949
[
50-
('fibroblast-prolif/prolif', 'fibroblast/prolif')
50+
('fibroblast-prolif/prolif', 'fibroblast/prolif'),
5151
]
5252
)
5353

5454
load_data(
5555
'https://bmcv.github.io/SuperDSM/fibroblast-ss.tar.bz2',
5656
[
57-
('fibroblast-ss/ss', 'fibroblast/ss')
57+
('fibroblast-ss/ss', 'fibroblast/ss'),
58+
]
59+
)
60+
61+
load_data(
62+
'https://data.broadinstitute.org/bbbc/BBBC039/images.zip',
63+
[
64+
('images', 'BBBC039'),
5865
]
5966
)

‎listfiles.sh

-5
This file was deleted.

0 commit comments

Comments
 (0)