Conda install incompatible with opencv for python 3.9

[nexplor]

🐛 Bug

Cannot install any version of torchvision newer than 0.2.2 with opencv for python 3.9

To Reproduce

Contents of ~/.condarc:

channels:
  - defaults
  - anaconda
  - pytorch
  - conda-forge
channel_priority: disabled

On the command line:

conda create -n temp python=3.9
conda activate temp
conda install torchvision opencv

This installs torchvision version 0.2.2.

Replacing the last line with

conda install torchvision=0.8 opencv

produces the error

UnsatisfiableError: The following specifications were found to be incompatible with each other:

Output in format: Requested package -> Available versions

Package openblas conflicts for:
opencv -> numpy[version='>=1.9.3,<2.0a0'] -> openblas[version='0.2.18.*|0.2.18|0.2.18.*|>=0.2.15']
opencv -> openblas[version='0.2.19|0.2.19.*|0.2.20|0.2.20.*|>=0.2.20,<0.2.21.0a0|>=0.3.3,<0.3.4.0a0|>=0.3.6,<0.3.7.0a0']

Package freetype conflicts for:
opencv -> libopencv==4.5.0=py37_4 -> freetype[version='2.6.*|2.7|2.7.*|2.8.1.*|2.8.1|>=2.10.4,<3.0a0|>=2.8.1,<2.9.0a0|>=2.10.2,<3.0a0|>=2.10.3,<3.0a0|>=2.8.1,<2.8.2.0a0']
opencv -> freetype[version='>=2.8,<2.9.0a0|>=2.9.1,<3.0a0']

Package _libgcc_mutex conflicts for:
opencv -> libgcc-ng[version='>=7.3.0'] -> _libgcc_mutex[version='*|0.1',build='main|conda_forge']
python==3.9 -> libgcc-ng[version='>=7.3.0'] -> _libgcc_mutex[version='*|0.1',build='main|conda_forge']

Package blas conflicts for:
opencv -> numpy[version='>=1.14.6,<2.0a0'] -> blas[version='*|*|1.0|1.0',build='openblas|openblas|mkl|mkl']
opencv -> blas==1.1=openblas

Package numpy conflicts for:
opencv -> py-opencv==4.5.0=py37h888b3d9_4 -> numpy[version='>=1.11.3,<2.0a0|>=1.16.5,<2.0a0|>=1.16.6,<2.0a0|>=1.19.4,<2.0a0|>=1.18.5,<2.0a0|>=1.19.2,<2.0a0|>=1.18.1,<2.0a0']
opencv -> numpy[version='1.10.*|1.11.*|1.12.*|1.13.*|>=1.14.6,<2.0a0|>=1.9.3,<2.0a0|>=1.8|1.9.*']

Package libtiff conflicts for:
opencv -> libtiff[version='4.0.*|>=4.0.10,<5.0a0|>=4.0.9,<5.0a0|>=4.0.8,<4.0.10|>=4.0.3,<4.0.8']
opencv -> libopencv==4.5.0=py37_4 -> libtiff[version='>=4.1.0,<5.0a0']

Package tzdata conflicts for:
torchvision=0.8 -> python[version='>=3.9,<3.10.0a0'] -> tzdata
python==3.9 -> tzdata

Package jpeg conflicts for:
opencv -> libopencv==4.5.0=py37_4 -> jpeg[version='9b|>=9d,<10a']
opencv -> jpeg[version='9.*|>=9c,<10a|>=9b,<10a']

Package numpy-base conflicts for:
torchvision=0.8 -> numpy[version='>=1.11'] -> numpy-base[version='1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.14.3|1.14.3|1.14.3|1.14.3|1.14.3|1.14.3|1.14.4|1.14.4|1.14.4|1.14.4|1.14.4|1.14.4|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.15.0|1.15.0|1.15.0|1.15.0|1.15.0|1.15.0|1.15.0|1.15.0|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.3|1.15.3|1.15.3|1.15.3|1.15.3|1.15.3|1.15.4|1.15.4|1.15.4|1.15.4|1.15.4|1.15.4|1.15.4|1.15.4|1.15.4|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.2|1.16.2|1.16.2|1.16.2|1.16.2|1.16.2|1.16.3|1.16.3|1.16.3|1.16.3|1.16.3|1.16.3|1.16.4|1.16.4|1.16.4|1.16.4|1.16.4|1.16.4|1.16.5|1.16.5|1.16.5|1.16.5|1.16.5|1.16.5|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.17.0|1.17.0|1.17.0|1.17.0|1.17.2.*|1.17.3.*|1.17.4.*|1.18.1.*|1.18.5.*|1.19.1|1.19.1|1.19.1|1.19.1|1.19.1|1.19.1|1.19.2',build='py27h2b20989_7|py36h2b20989_7|py36hdbf6ddf_8|py36h2b20989_8|py37hdbf6ddf_8|py37h2b20989_8|py27h2b20989_8|py27hdbf6ddf_8|py35hdbf6ddf_8|py35h7cdd4dd_9|py27h7cdd4dd_9|py37h3dfced4_9|py27h3dfced4_9|py37h81de0dd_9|py35h74e8950_9|py37h74e8950_9|py36h74e8950_9|py35h81de0dd_9|py37h74e8950_10|py35h74e8950_10|py36h74e8950_10|py37h81de0dd_10|py36h81de0dd_10|py36h2f8d375_10|py27h2f8d375_10|py37h2f8d375_11|py27h2f8d375_11|py36h2f8d375_11|py37hde5b4d6_11|py27hde5b4d6_11|py27hde5b4d6_12|py38hde5b4d6_12|py38h2f8d375_12|py27h9be14a7_1|py27h2b20989_0|py36h2b20989_0|py35h2b20989_0|py27hdbf6ddf_0|py36hdbf6ddf_0|py36h2b20989_0|py27h2b20989_0|py36hdbf6ddf_0|py35hdbf6ddf_0|py36h2b20989_1|py27hdbf6ddf_1|py36hdbf6ddf_1|py27h2b20989_1|py27h2b20989_2|py36h2b20989_2|py36hdbf6ddf_3|py27h2b20989_3|py27hdbf6ddf_4|py37h2b20989_4|py37hdbf6ddf_4|py36h2b20989_4|py37h2f8d375_4|py35h2f8d375_4|py27h81de0dd_4|py36h2f8d375_4|py35h81de0dd_4|py37h81de0dd_4|py38hde5b4d6_4|py37h2f8d375_5|py36hde5b4d6_5|py27h7cdd4dd_0|py35h3dfced4_0|py37h3dfced4_0|py27h74e8950_0|py36h74e8950_0|py36h81de0dd_0|py27h81de0dd_0|py35h2f8d375_0|py37h2f8d375_0|py27h2f8d375_0|py35h2f8d375_0|py36h81de0dd_0|py27h2f8d375_1|py27h81de0dd_1|py36h81de0dd_1|py37h2f8d375_1|py27h2f8d375_0|py36h2f8d375_0|py36h81de0dd_0|py36h81de0dd_0|py27h2f8d375_0|py27hde5b4d6_0|py36hde5b4d6_0|py27h2f8d375_1|py36h2f8d375_1|py27hde5b4d6_1|py36hde5b4d6_1|py37h2f8d375_0|py36h2f8d375_0|py27hde5b4d6_0|py37h2f8d375_1|py37hde5b4d6_1|py36hde5b4d6_1|py27hde5b4d6_1|py37hde5b4d6_0|py27h2f8d375_0|py27hde5b4d6_0|py36hde5b4d6_0|py36hde5b4d6_0|py37h2f8d375_0|py37hfa32c7d_0|py38h75fe3a5_0|py38hfa32c7d_0|py36h75fe3a5_0|py37hfa32c7d_0|py38h75fe3a5_0|py36h75fe3a5_0|py38hfa32c7d_0|py39h2ae0177_0|py39h0f7b65f_0|py36hfa32c7d_0|py37h75fe3a5_0|py37h75fe3a5_0|py36hfa32c7d_0|py36hde5b4d6_0|py37hde5b4d6_0|py37h2f8d375_0|py36h2f8d375_0|py39h76555f2_1|py39hfb011de_1|py27hde5b4d6_0|py27h2f8d375_0|py38hde5b4d6_0|py38h2f8d375_0|py36hde5b4d6_0|py37hde5b4d6_0|py36h2f8d375_0|py37h2f8d375_0|py36hde5b4d6_0|py37hde5b4d6_0|py27hde5b4d6_0|py27h2f8d375_0|py36h2f8d375_0|py27hde5b4d6_0|py37hde5b4d6_0|py37h2f8d375_0|py36h2f8d375_0|py27h2f8d375_0|py37hde5b4d6_0|py36h2f8d375_0|py37h2f8d375_0|py36hde5b4d6_0|py27hde5b4d6_0|py27h2f8d375_0|py36h2f8d375_0|py37h2f8d375_0|py36h2f8d375_1|py27h2f8d375_1|py37hde5b4d6_0|py36hde5b4d6_0|py27h2f8d375_0|py37hde5b4d6_1|py37h2f8d375_1|py36h2f8d375_0|py37hde5b4d6_0|py37h2f8d375_0|py37hde5b4d6_0|py36hde5b4d6_0|py27hde5b4d6_0|py27h81de0dd_0|py37h81de0dd_0|py27h2f8d375_0|py36h2f8d375_0|py37h2f8d375_0|py37h81de0dd_0|py27h81de0dd_0|py37h2f8d375_0|py37h81de0dd_1|py36h2f8d375_1|py36h2f8d375_0|py35h81de0dd_0|py27h81de0dd_0|py37h81de0dd_0|py37h2f8d375_0|py36h2f8d375_0|py27h2f8d375_0|py35h81de0dd_0|py37h81de0dd_0|py37h74e8950_0|py35h74e8950_0|py36h3dfced4_0|py27h3dfced4_0|py35h7cdd4dd_0|py37h7cdd4dd_0|py36h7cdd4dd_0|py37hde5b4d6_5|py27hde5b4d6_5|py27h2f8d375_5|py36h2f8d375_5|py38h2f8d375_4|py36h81de0dd_4|py27h2f8d375_4|py35h2b20989_4|py35hdbf6ddf_4|py36hdbf6ddf_4|py27h2b20989_4|py27hdbf6ddf_3|py37hdbf6ddf_3|py37h2b20989_3|py36h2b20989_3|py36hdbf6ddf_2|py37hdbf6ddf_2|py27hdbf6ddf_2|py37h2b20989_2|py37hdbf6ddf_1|py37h2b20989_1|py27hdbf6ddf_0|py35hdbf6ddf_0|py35h9be14a7_1|py36h9be14a7_1|py35h0ea5e3f_1|py27h0ea5e3f_1|py36h0ea5e3f_1|py36hde5b4d6_12|py37hde5b4d6_12|py36h2f8d375_12|py27h2f8d375_12|py37h2f8d375_12|py36hde5b4d6_11|py35h2f8d375_10|py37h2f8d375_10|py35h81de0dd_10|py27h81de0dd_10|py27h74e8950_10|py27h81de0dd_9|py27h74e8950_9|py36h81de0dd_9|py35h3dfced4_9|py36h3dfced4_9|py37h7cdd4dd_9|py36h7cdd4dd_9|py35h2b20989_8|py37hdbf6ddf_7|py36hdbf6ddf_7|py27hdbf6ddf_7|py37h2b20989_7']
opencv -> numpy[version='>=1.14.6,<2.0a0'] -> numpy-base[version='1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.11.3|1.14.3|1.14.3|1.14.3|1.14.3|1.14.3|1.14.3|1.14.4|1.14.4|1.14.4|1.14.4|1.14.4|1.14.4|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.5|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.14.6|1.15.0|1.15.0|1.15.0|1.15.0|1.15.0|1.15.0|1.15.0|1.15.0|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.1|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.2|1.15.3|1.15.3|1.15.3|1.15.3|1.15.3|1.15.3|1.15.4|1.15.4|1.15.4|1.15.4|1.15.4|1.15.4|1.15.4|1.15.4|1.15.4|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.0|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.1|1.16.2|1.16.2|1.16.2|1.16.2|1.16.2|1.16.2|1.16.3|1.16.3|1.16.3|1.16.3|1.16.3|1.16.3|1.16.4|1.16.4|1.16.4|1.16.4|1.16.4|1.16.4|1.16.5|1.16.5|1.16.5|1.16.5|1.16.5|1.16.5|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.16.6|1.17.0|1.17.0|1.17.0|1.17.0|1.17.2.*|1.17.3.*|1.17.4.*|1.18.1.*|1.18.5.*|1.19.1|1.19.1|1.19.1|1.19.1|1.19.1|1.19.1|1.19.2|1.9.3|1.9.3|1.9.3|1.9.3|1.9.3|1.9.3|1.9.3|1.9.3|1.9.3|1.9.3|1.9.3|1.9.3|1.9.3|1.9.3|>=1.9.3,<2.0a0',build='py37h2b20989_6|py37hdbf6ddf_6|py27h2b20989_7|py36h2b20989_7|py35hdbf6ddf_7|py27h2b20989_7|py36h2b20989_7|py36hdbf6ddf_8|py36h2b20989_8|py37hdbf6ddf_8|py37h2b20989_8|py27h2b20989_8|py27hdbf6ddf_8|py35hdbf6ddf_8|py35h7cdd4dd_9|py27h7cdd4dd_9|py37h3dfced4_9|py27h3dfced4_9|py37h81de0dd_9|py35h74e8950_9|py37h74e8950_9|py36h74e8950_9|py35h81de0dd_9|py37h74e8950_10|py35h74e8950_10|py36h74e8950_10|py37h81de0dd_10|py36h81de0dd_10|py36h2f8d375_10|py27h2f8d375_10|py37h2f8d375_11|py27h2f8d375_11|py36h2f8d375_11|py37hde5b4d6_11|py27hde5b4d6_11|py27hde5b4d6_12|py38hde5b4d6_12|py38h2f8d375_12|py27h9be14a7_1|py27h2b20989_0|py36h2b20989_0|py35h2b20989_0|py27hdbf6ddf_0|py36hdbf6ddf_0|py36h2b20989_0|py27h2b20989_0|py36hdbf6ddf_0|py35hdbf6ddf_0|py36h2b20989_1|py27hdbf6ddf_1|py36hdbf6ddf_1|py27h2b20989_1|py27h2b20989_2|py36h2b20989_2|py36hdbf6ddf_3|py27h2b20989_3|py27hdbf6ddf_4|py37h2b20989_4|py37hdbf6ddf_4|py36h2b20989_4|py37h2f8d375_4|py35h2f8d375_4|py27h81de0dd_4|py36h2f8d375_4|py35h81de0dd_4|py37h81de0dd_4|py38hde5b4d6_4|py37h2f8d375_5|py36hde5b4d6_5|py27h7cdd4dd_0|py35h3dfced4_0|py37h3dfced4_0|py27h74e8950_0|py36h74e8950_0|py36h81de0dd_0|py27h81de0dd_0|py35h2f8d375_0|py37h2f8d375_0|py27h2f8d375_0|py35h2f8d375_0|py36h81de0dd_0|py27h2f8d375_1|py27h81de0dd_1|py36h81de0dd_1|py37h2f8d375_1|py27h2f8d375_0|py36h2f8d375_0|py36h81de0dd_0|py36h81de0dd_0|py27h2f8d375_0|py27hde5b4d6_0|py36hde5b4d6_0|py27h2f8d375_1|py36h2f8d375_1|py27hde5b4d6_1|py36hde5b4d6_1|py37h2f8d375_0|py36h2f8d375_0|py27hde5b4d6_0|py37h2f8d375_1|py37hde5b4d6_1|py36hde5b4d6_1|py27hde5b4d6_1|py37hde5b4d6_0|py27h2f8d375_0|py27hde5b4d6_0|py36hde5b4d6_0|py36hde5b4d6_0|py37h2f8d375_0|py37hfa32c7d_0|py38h75fe3a5_0|py38hfa32c7d_0|py36h75fe3a5_0|py37hfa32c7d_0|py38h75fe3a5_0|py36h75fe3a5_0|py38hfa32c7d_0|py39h2ae0177_0|py39h0f7b65f_0|py36hfa32c7d_0|py37h75fe3a5_0|py37h75fe3a5_0|py36hfa32c7d_0|py36hde5b4d6_0|py37hde5b4d6_0|py37h2f8d375_0|py36h2f8d375_0|py39h76555f2_1|py39hfb011de_1|py27hde5b4d6_0|py27h2f8d375_0|py38hde5b4d6_0|py38h2f8d375_0|py36hde5b4d6_0|py37hde5b4d6_0|py36h2f8d375_0|py37h2f8d375_0|py36hde5b4d6_0|py37hde5b4d6_0|py27hde5b4d6_0|py27h2f8d375_0|py36h2f8d375_0|py27hde5b4d6_0|py37hde5b4d6_0|py37h2f8d375_0|py36h2f8d375_0|py27h2f8d375_0|py37hde5b4d6_0|py36h2f8d375_0|py37h2f8d375_0|py36hde5b4d6_0|py27hde5b4d6_0|py27h2f8d375_0|py36h2f8d375_0|py37h2f8d375_0|py36h2f8d375_1|py27h2f8d375_1|py37hde5b4d6_0|py36hde5b4d6_0|py27h2f8d375_0|py37hde5b4d6_1|py37h2f8d375_1|py36h2f8d375_0|py37hde5b4d6_0|py37h2f8d375_0|py37hde5b4d6_0|py36hde5b4d6_0|py27hde5b4d6_0|py27h81de0dd_0|py37h81de0dd_0|py27h2f8d375_0|py36h2f8d375_0|py37h2f8d375_0|py37h81de0dd_0|py27h81de0dd_0|py37h2f8d375_0|py37h81de0dd_1|py36h2f8d375_1|py36h2f8d375_0|py35h81de0dd_0|py27h81de0dd_0|py37h81de0dd_0|py37h2f8d375_0|py36h2f8d375_0|py27h2f8d375_0|py35h81de0dd_0|py37h81de0dd_0|py37h74e8950_0|py35h74e8950_0|py36h3dfced4_0|py27h3dfced4_0|py35h7cdd4dd_0|py37h7cdd4dd_0|py36h7cdd4dd_0|py37hde5b4d6_5|py27hde5b4d6_5|py27h2f8d375_5|py36h2f8d375_5|py38h2f8d375_4|py36h81de0dd_4|py27h2f8d375_4|py35h2b20989_4|py35hdbf6ddf_4|py36hdbf6ddf_4|py27h2b20989_4|py27hdbf6ddf_3|py37hdbf6ddf_3|py37h2b20989_3|py36h2b20989_3|py36hdbf6ddf_2|py37hdbf6ddf_2|py27hdbf6ddf_2|py37h2b20989_2|py37hdbf6ddf_1|py37h2b20989_1|py27hdbf6ddf_0|py35hdbf6ddf_0|py35h9be14a7_1|py36h9be14a7_1|py35h0ea5e3f_1|py27h0ea5e3f_1|py36h0ea5e3f_1|py36hde5b4d6_12|py37hde5b4d6_12|py36h2f8d375_12|py27h2f8d375_12|py37h2f8d375_12|py36hde5b4d6_11|py35h2f8d375_10|py37h2f8d375_10|py35h81de0dd_10|py27h81de0dd_10|py27h74e8950_10|py27h81de0dd_9|py27h74e8950_9|py36h81de0dd_9|py35h3dfced4_9|py36h3dfced4_9|py37h7cdd4dd_9|py36h7cdd4dd_9|py35h2b20989_8|py37hdbf6ddf_7|py36hdbf6ddf_7|py27hdbf6ddf_7|py37h2b20989_7|py37h2b20989_7|py37hdbf6ddf_7|py35h2b20989_7|py36hdbf6ddf_7|py27hdbf6ddf_7|py36hdbf6ddf_6|py27hdbf6ddf_6|py36h2b20989_6|py27h2b20989_6']The following specifications were found to be incompatible with your system:

  - feature:/linux-64::__glibc==2.32=0
  - feature:|@/linux-64::__glibc==2.32=0
  - torchvision=0.8 -> cudatoolkit[version='>=11.0,<11.1'] -> __glibc[version='>=2.17,<3.0.a0']

Your installed version is: 2.32

Expected behavior

Torchvision version 0.8.2 should be selected for installation.
If version is explicitly specified, packages should get installed without error.

Environment

OS: Ubuntu 20.10 (x86_64)
GCC version: (Ubuntu 10.2.0-13ubuntu1) 10.2.0
Clang version: 11.0.0-2
CMake version: version 3.16.3
Python version: 3.9 (64-bit runtime)
GPU models and configuration: GPU 0: GeForce 930MX
Nvidia driver version: 450.80.02

[farleylai]

jpeg now requires <= 9b which is very old unlike previous releases.
Other packages such as the popular opencv already requires jpeg >=9d for long since v4.2.
This leads to conflicts clearly.
So what is the specific reason for that?

[ManifoldFR]

There are similar problems for Python 3.8.

I also saw that installing a working opencv and PyTorch GPU (from the pytorch channel) with conda is impossible - conda will install pytorch::ffmpeg 4.3.0 which leads to errors when doing import cv2. Also installing torchvision leads to even more problems.

[mrgransky]

Do you also get the same error downloading it via pip in your conda env?

$ pip install opencv-contrib-python

I had the same problem with python 3.9.1, but I installed opencv with pip and it works with

$ python -c "import cv2; print(cv2.__version__)"
4.5.1

[astromid]

Have the same problem, can't install latest torchvision releases (which requires jpeg <= 9b) together with opencv (requires jpeg >= 9d) in conda environment.
My workaround for this is to install anything by conda (except torchvision) and at the end install corresponding torchvision build with pip install --no-deps
Are there any plans to support jpeg >= 9d in future torchvision releases?

[cattaneod]

Same issue here, installing opencv using pip as @mrgransky said solved.

[farleylai]

As of torchvision-0.9.1, the dependency issue remains, blocking the installation of opencv through conda:

$ conda search --info torchvision=0.9.1
torchvision 0.9.1 py37_cu111
----------------------------
file name   : torchvision-0.9.1-py37_cu111.tar.bz2
name        : torchvision
version     : 0.9.1
build       : py37_cu111
build number: 0
size        : 25.9 MB
license     : BSD
subdir      : linux-64
url         : https://conda.anaconda.org/pytorch/linux-64/torchvision-0.9.1-py37_cu111.tar.bz2
md5         : d47215411c0bc1ca792d141584ec341a
timestamp   : 2021-03-25 04:11:13 UTC
dependencies: 
  - cudatoolkit >=11.1,<11.2
  - ffmpeg >=4.2
  - jpeg <=9b
  - libpng
  - numpy >=1.11
  - pillow >=4.1.1
  - python >=3.7,<3.8.0a0
  - pytorch 1.8.1

Thus the upgrade to pytorch-1.8.1 is still unlikely.
Sticking to pytorch-1.7.0 seems inevitable for no specific jpeg version requirement:

$ conda search --info torchvision=0.8.1
torchvision 0.8.1 py38_cu110
----------------------------
file name   : torchvision-0.8.1-py38_cu110.tar.bz2
name        : torchvision
version     : 0.8.1
build       : py38_cu110
build number: 0
size        : 17.9 MB
license     : BSD
subdir      : linux-64
url         : https://conda.anaconda.org/pytorch/linux-64/torchvision-0.8.1-py38_cu110.tar.bz2
md5         : 904400aebba9b6b5a485d36931205324
timestamp   : 2020-10-27 20:39:32 UTC
dependencies: 
  - cudatoolkit >=11.0,<11.1
  - jpeg
  - libpng
  - numpy >=1.11
  - pillow >=4.1.1
  - python >=3.8,<3.9.0a0
  - pytorch 1.7.0

Update: torchvision built by conda-forge does NOT have this old jpeg<=9b dependency.

torchvision 0.9.0 py37cuda110ha9b8171_0_cuda
--------------------------------------------
file name   : torchvision-0.9.0-py37cuda110ha9b8171_0_cuda.tar.bz2
name        : torchvision
version     : 0.9.0
build       : py37cuda110ha9b8171_0_cuda
build number: 0
size        : 7.6 MB
license     : BSD-3-Clause
subdir      : linux-64
url         : https://conda.anaconda.org/conda-forge/linux-64/torchvision-0.9.0-py37cuda110ha9b8171_0_cuda.tar.bz2
md5         : f97ced63c67c7fbd4a0d43096df2179d
timestamp   : 2021-03-16 18:39:22 UTC
dependencies: 
  - __glibc >=2.17
  - __glibc >=2.17,<3.0.a0
  - cudatoolkit 11.0|11.0.*
  - cudnn >=8.1.0.77,<9.0a0
  - jpeg >=9d,<10a
  - libgcc-ng >=7.5.0
  - libpng >=1.6.37,<1.7.0a0
  - libstdcxx-ng >=7.5.0
  - numpy >=1.16.6,<2.0a0
  - pillow >=4.1.1
  - python >=3.7,<3.8.0a0
  - python_abi 3.7.* *_cp37m
  - pytorch >=1.8.0 cuda*

So, jpeg <=9b was checked in since commit 2857536 cherry picked from
2f40a48 pinned by @seemethere in #3063. Any justifications?

[fmassa]

Pinging @seemethere and @andfoy as this looks like a conflict which would be good to address

[seemethere]

We can try without a pinned version of jpeg, the previous pinning was due to a dependency issue I was seeing back when we first added Python 3.9, perhaps the issue has resolved itself by this point