forked from ilastik/ilastik
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathenvironment-dev.yml
87 lines (81 loc) · 2.58 KB
/
environment-dev.yml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
channels:
- pytorch
# - nvidia # uncomment for pytorch with cuda
# ilastik-forge/label/patched contains patched versions of pyshtools
# with certain dependencies removed that made solving the ilastik
# environment impossible / extremely inflexible
- ilastik-forge/label/patched
- ilastik-forge
- conda-forge
- nodefaults
dependencies:
- python 3.9.*
- numpy 1.22.*
- appdirs
- cachetools
- dpct
- fastfilters
- future
- greenlet
- grpcio 1.49.1
- h5py
- hytra
- ilastik-feature-selection
- ilastikrag >=0.1.4
- ilastiktools
- jsonschema
- mamutexport
- marching_cubes
- ndstructs
- nifty
- pandas 2.*
- psutil
- pydantic 2.*
- pyopengl
- pyqt 5.15.*
- pyqtgraph
# previous versions would set thread limits globally with side effects
- python-elf >= 0.4.8
- qimage2ndarray
- scikit-image
- scikit-learn
# for python 3.7 compatible environment use
# tiffile >2020.9.22,<=2021.11.2
- tifffile >=2022
# build 1.11.1=*_1028 on cf is the first to be compatible with numpy>1.19
# need to bump this manually until there is a true version bump in vigra
- vigra 1.11.1=*_1038
# xarray versions not compatible with numpy 1.21, 2023.08.0 might be, but lost trust
# 2023.10.1 correctly pins numpy to 1.22 and up
- xarray !=2023.8.0,!=2023.9.0,!=2023.10.0
- z5py
- zarr 2.*
- aiohttp
- fsspec
- s3fs >=2022.8.2
# Deep learning dependencies (neural net workflow, trainable domain adaptation)
# can be changed to request gpu versions
- pytorch <2.4.1 # >=2.4 can clash with pyshtools on win; 2.2.2 newest available on osx
- cpuonly # comment out for pytorch with cuda
# - pytorch-cuda=11.3 # uncomment for pytorch with cuda, specify cuda version
- tiktorch 24.12.0
# mamba does not "respect" track_features, which is the idea behind cpuonly,
# with ilastik-pytorch-version-helper-cpu we help mamba on linux and windows
# to install a cpu-build
- ilastik-pytorch-version-helper-cpu # comment out for pytorch with cuda
# Third-party object feature plugins
- sphericaltexture_plugin
# TODO: sphericaltexture 0.1.1 currently times out in the tests
- sphericaltexture 0.0.4
- pyshtools >=4.12 # If >=4.13 becomes available for win, might be able to unpin pytorch
# dev-only dependencies
- conda-build
- mypy
- pre-commit
- pytest >4
- pytest-qt
# ensure working environment on apple M1 via rosetta
# remove once native builds are available
# - mkl 2021.* # [osx]
# must be commented out on arm-64 native (no mkl!)
- mkl <2024.1.0 # [linux] until pytorch is compatible with the current version