search

astral-sh / uv

An extremely fast Python package and project manager, written in Rust.
https://docs.astral.sh/uv
Apache License 2.0
26.29k stars 766 forks source link

Bad resolver error for `colabfold[alphafold]==1.5.5` on python 3.11 #5046

Closed konstin closed 2 months ago

konstin commented 4 months ago

The resolver error for echo "colabfold[alphafold]==1.5.5" | cargo run pip compile -p 3.11 --universal - is verbose and has multiple problems. I'm filing this as one ticket rather than multiple since i expect that some of this is #5045 and others may be the same underlying problem. We should split it into multiple after tracking down the cause of the different problems.

Full resolver error ``` × No solution found when resolving dependencies: ╰─▶ Because jax>=0.4.20 depends on numpy{python_version >= '3.12'}>=1.26.0 and only the following versions of jax are available: jax<=0.4.20 jax==0.4.21 jax==0.4.22 jax==0.4.23 jax==0.4.24 jax==0.4.25 jax==0.4.26 jax==0.4.27 jax==0.4.28 jax==0.4.29 jax==0.4.30 jax>=0.5.0 we can conclude that jax>=0.4.20 depends on numpy{python_version >= '3.12'}>=1.26.0. And because colabfold[alphafold]==1.5.5 depends on jax>=0.4.20 and only the following versions of numpy{python_version >= '3.12'} are available: numpy{python_version >= '3.12'}<=1.26.0 numpy{python_version >= '3.12'}==1.26.1 numpy{python_version >= '3.12'}==1.26.2 numpy{python_version >= '3.12'}==1.26.3 numpy{python_version >= '3.12'}==1.26.4 numpy{python_version >= '3.12'}==2.0.0 we can conclude that colabfold[alphafold]==1.5.5, any of: numpy<1.26.0 numpy>1.26.0,<1.26.1 numpy>1.26.1,<1.26.2 numpy>1.26.2,<1.26.3 numpy>1.26.3,<2.0.0 numpy>2.0.0 , numpy{python_version >= '3.12'}!=1.26.4 are incompatible. (1) Because only the following versions of tensorflow-cpu{sys_platform != 'darwin'} are available: tensorflow-cpu{sys_platform != 'darwin'}<=2.12.1 tensorflow-cpu{sys_platform != 'darwin'}==2.13.0 tensorflow-cpu{sys_platform != 'darwin'}==2.13.1 tensorflow-cpu{sys_platform != 'darwin'}==2.14.0 tensorflow-cpu{sys_platform != 'darwin'}==2.14.1 tensorflow-cpu{sys_platform != 'darwin'}==2.15.0 tensorflow-cpu{sys_platform != 'darwin'}==2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}==2.15.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.2 tensorflow-cpu{sys_platform != 'darwin'}==2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>=3.0.0 and any of: tensorflow-cpu{sys_platform != 'darwin'}==2.12.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.13.0,<=2.13.1 depend on numpy>=1.22,<=1.24.3, we can conclude that any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.13.0 tensorflow-cpu{sys_platform != 'darwin'}>2.13.0,<2.13.1 tensorflow-cpu{sys_platform != 'darwin'}>2.13.1,<2.14.0 tensorflow-cpu{sys_platform != 'darwin'}>2.14.0,<2.14.1 tensorflow-cpu{sys_platform != 'darwin'}>2.14.1,<2.15.0 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0,<2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0.post1,<2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 depend on numpy>=1.22,<=1.24.3. And because tensorflow-cpu{sys_platform != 'darwin'}>=2.13.0,<=2.13.1 depends on numpy>=1.22,<=1.24.3 and numpy>=1.22,<=1.24.3, we can conclude that any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.14.0 tensorflow-cpu{sys_platform != 'darwin'}>2.14.0,<2.14.1 tensorflow-cpu{sys_platform != 'darwin'}>2.14.1,<2.15.0 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0,<2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0.post1,<2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 depend on numpy>=1.22,<=1.24.3. (2) Because tensorflow-macos{sys_platform == 'darwin'}==2.10.0 depends on tensorflow-macos{sys_platform == 'darwin'}==2.10.0 and tensorflow-macos{sys_platform == 'darwin'}>=2.9.1,<=2.11.0 depends on protobuf>=3.9.2,<3.20, we can conclude that any of: tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 depend on protobuf>=3.9.2,<3.20. And because tensorflow-macos{sys_platform == 'darwin'}==2.8.0 depends on keras>=2.8.0rc0,<2.9 and flatbuffers>=1.12,<2, we can conclude that any of: flatbuffers<1.12 flatbuffers>=2 , any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , tensorflow-macos{sys_platform == 'darwin'}!=2.7.0, any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}==2.7.0 depends on flatbuffers>=1.12,<3.0 and any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.14.0,<=2.14.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.15.0,<=2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 depend on flatbuffers>=23.5.26, we can conclude that any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , any of: tensorflow-cpu{sys_platform != 'darwin'}==2.14.0 tensorflow-cpu{sys_platform != 'darwin'}==2.14.1 tensorflow-cpu{sys_platform != 'darwin'}==2.15.0 tensorflow-cpu{sys_platform != 'darwin'}==2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}==2.15.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.2 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.14.0,<=2.14.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.15.0,<=2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>=2.17.0 depend on one of: protobuf>=3.20.3,<4.21.0 protobuf>4.21.0,<4.21.1 protobuf>4.21.1,<4.21.2 protobuf>4.21.2,<4.21.3 protobuf>4.21.3,<4.21.4 protobuf>4.21.4,<4.21.5 protobuf>4.21.5,<5.0.0.dev0 and keras>=2.14.0,<2.15, we can conclude that any of: tensorflow-cpu{sys_platform != 'darwin'}==2.14.0 tensorflow-cpu{sys_platform != 'darwin'}==2.14.1 and any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because we know from (2) that any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.14.0 tensorflow-cpu{sys_platform != 'darwin'}>2.14.0,<2.14.1 tensorflow-cpu{sys_platform != 'darwin'}>2.14.1,<2.15.0 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0,<2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0.post1,<2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 depend on numpy>=1.22,<=1.24.3, we can conclude that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.14.1 tensorflow-cpu{sys_platform != 'darwin'}>2.14.1,<2.15.0 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0,<2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0.post1,<2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. (3) Because tensorflow-macos{sys_platform == 'darwin'}==2.10.0 depends on tensorflow-macos{sys_platform == 'darwin'}==2.10.0 and tensorflow-macos{sys_platform == 'darwin'}>=2.9.1,<=2.11.0 depends on protobuf>=3.9.2,<3.20, we can conclude that any of: tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 depend on protobuf>=3.9.2,<3.20. And because tensorflow-macos{sys_platform == 'darwin'}==2.8.0 depends on keras>=2.8.0rc0,<2.9 and flatbuffers>=1.12,<2, we can conclude that any of: flatbuffers<1.12 flatbuffers>=2 , any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , tensorflow-macos{sys_platform == 'darwin'}!=2.7.0, any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}==2.7.0 depends on flatbuffers>=1.12,<3.0 and any of: tensorflow-cpu{sys_platform != 'darwin'}==2.14.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.15.0,<=2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 depend on flatbuffers>=23.5.26, we can conclude that any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , any of: tensorflow-cpu{sys_platform != 'darwin'}==2.14.1 tensorflow-cpu{sys_platform != 'darwin'}==2.15.0 tensorflow-cpu{sys_platform != 'darwin'}==2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}==2.15.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.2 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because any of: tensorflow-cpu{sys_platform != 'darwin'}==2.14.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.15.0,<=2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>=2.17.0 depend on one of: protobuf>=3.20.3,<4.21.0 protobuf>4.21.0,<4.21.1 protobuf>4.21.1,<4.21.2 protobuf>4.21.2,<4.21.3 protobuf>4.21.3,<4.21.4 protobuf>4.21.4,<4.21.5 protobuf>4.21.5,<5.0.0.dev0 and keras>=2.14.0,<2.15, we can conclude that tensorflow-cpu{sys_platform != 'darwin'}==2.14.1 and any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because we know from (3) that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.14.1 tensorflow-cpu{sys_platform != 'darwin'}>2.14.1,<2.15.0 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0,<2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0.post1,<2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible, we can conclude that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.15.0 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0,<2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0.post1,<2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. (4) Because tensorflow-macos{sys_platform == 'darwin'}==2.10.0 depends on tensorflow-macos{sys_platform == 'darwin'}==2.10.0 and tensorflow-macos{sys_platform == 'darwin'}>=2.9.1,<=2.11.0 depends on protobuf>=3.9.2,<3.20, we can conclude that any of: tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 depend on protobuf>=3.9.2,<3.20. And because tensorflow-macos{sys_platform == 'darwin'}==2.8.0 depends on keras>=2.8.0rc0,<2.9 and flatbuffers>=1.12,<2, we can conclude that any of: flatbuffers<1.12 flatbuffers>=2 , any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , tensorflow-macos{sys_platform == 'darwin'}!=2.7.0, any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}==2.7.0 depends on flatbuffers>=1.12,<3.0 and any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.15.0,<=2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 depend on flatbuffers>=23.5.26, we can conclude that any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , any of: tensorflow-cpu{sys_platform != 'darwin'}==2.15.0 tensorflow-cpu{sys_platform != 'darwin'}==2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}==2.15.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.2 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.15.0,<=2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>=2.17.0 depend on one of: protobuf>=3.20.3,<4.21.0 protobuf>4.21.0,<4.21.1 protobuf>4.21.1,<4.21.2 protobuf>4.21.2,<4.21.3 protobuf>4.21.3,<4.21.4 protobuf>4.21.4,<4.21.5 protobuf>4.21.5,<5.0.0.dev0 and keras>=2.15.0,<2.16, we can conclude that any of: tensorflow-cpu{sys_platform != 'darwin'}==2.15.0 tensorflow-cpu{sys_platform != 'darwin'}==2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}==2.15.1 and any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because we know from (4) that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.15.0 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0,<2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0.post1,<2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible, we can conclude that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0.post1,<2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. (5) Because tensorflow-macos{sys_platform == 'darwin'}==2.10.0 depends on tensorflow-macos{sys_platform == 'darwin'}==2.10.0 and tensorflow-macos{sys_platform == 'darwin'}>=2.9.1,<=2.11.0 depends on protobuf>=3.9.2,<3.20, we can conclude that any of: tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 depend on protobuf>=3.9.2,<3.20. And because tensorflow-macos{sys_platform == 'darwin'}==2.8.0 depends on keras>=2.8.0rc0,<2.9 and flatbuffers>=1.12,<2, we can conclude that any of: flatbuffers<1.12 flatbuffers>=2 , any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , tensorflow-macos{sys_platform == 'darwin'}!=2.7.0, any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}==2.7.0 depends on flatbuffers>=1.12,<3.0 and any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.15.0.post1,<=2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 depend on flatbuffers>=23.5.26, we can conclude that any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , any of: tensorflow-cpu{sys_platform != 'darwin'}==2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}==2.15.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.2 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.15.0.post1,<=2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>=2.17.0 depend on one of: protobuf>=3.20.3,<4.21.0 protobuf>4.21.0,<4.21.1 protobuf>4.21.1,<4.21.2 protobuf>4.21.2,<4.21.3 protobuf>4.21.3,<4.21.4 protobuf>4.21.4,<4.21.5 protobuf>4.21.5,<5.0.0.dev0 and keras>=2.15.0,<2.16, we can conclude that any of: tensorflow-cpu{sys_platform != 'darwin'}==2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}==2.15.1 and any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because we know from (5) that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.15.0.post1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.0.post1,<2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible, we can conclude that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. (6) Because tensorflow-macos{sys_platform == 'darwin'}==2.10.0 depends on tensorflow-macos{sys_platform == 'darwin'}==2.10.0 and tensorflow-macos{sys_platform == 'darwin'}>=2.9.1,<=2.11.0 depends on protobuf>=3.9.2,<3.20, we can conclude that any of: tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 depend on protobuf>=3.9.2,<3.20. And because tensorflow-macos{sys_platform == 'darwin'}==2.8.0 depends on keras>=2.8.0rc0,<2.9 and flatbuffers>=1.12,<2, we can conclude that any of: flatbuffers<1.12 flatbuffers>=2 , any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , tensorflow-macos{sys_platform == 'darwin'}!=2.7.0, any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}==2.7.0 depends on flatbuffers>=1.12,<3.0 and any of: tensorflow-cpu{sys_platform != 'darwin'}==2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 depend on flatbuffers>=23.5.26, we can conclude that any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , any of: tensorflow-cpu{sys_platform != 'darwin'}==2.15.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.2 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because any of: tensorflow-cpu{sys_platform != 'darwin'}==2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>=2.17.0 depend on one of: protobuf>=3.20.3,<4.21.0 protobuf>4.21.0,<4.21.1 protobuf>4.21.1,<4.21.2 protobuf>4.21.2,<4.21.3 protobuf>4.21.3,<4.21.4 protobuf>4.21.4,<4.21.5 protobuf>4.21.5,<5.0.0.dev0 and keras>=2.15.0,<2.16, we can conclude that tensorflow-cpu{sys_platform != 'darwin'}==2.15.1 and any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because we know from (6) that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.15.1 tensorflow-cpu{sys_platform != 'darwin'}>2.15.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible, we can conclude that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. (7) Because tensorflow-macos{sys_platform == 'darwin'}==2.10.0 depends on tensorflow-macos{sys_platform == 'darwin'}==2.10.0 and tensorflow-macos{sys_platform == 'darwin'}>=2.9.1,<=2.11.0 depends on protobuf>=3.9.2,<3.20, we can conclude that any of: tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 depend on protobuf>=3.9.2,<3.20. And because tensorflow-macos{sys_platform == 'darwin'}==2.8.0 depends on keras>=2.8.0rc0,<2.9 and flatbuffers>=1.12,<2, we can conclude that any of: flatbuffers<1.12 flatbuffers>=2 , any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , tensorflow-macos{sys_platform == 'darwin'}!=2.7.0, any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}==2.7.0 depends on flatbuffers>=1.12,<3.0 and tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 depends on flatbuffers>=23.5.26, we can conclude that any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , any of: tensorflow-cpu{sys_platform != 'darwin'}==2.16.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.2 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.16.1,<=2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>=2.17.0 depend on one of: protobuf>=3.20.3,<4.21.0 protobuf>4.21.0,<4.21.1 protobuf>4.21.1,<4.21.2 protobuf>4.21.2,<4.21.3 protobuf>4.21.3,<4.21.4 protobuf>4.21.4,<4.21.5 protobuf>4.21.5,<5.0.0.dev0 and keras>=3.0.0, we can conclude that any of: tensorflow-cpu{sys_platform != 'darwin'}==2.16.1 tensorflow-cpu{sys_platform != 'darwin'}==2.16.2 and any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because we know from (7) that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.16.1 tensorflow-cpu{sys_platform != 'darwin'}>2.16.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible, we can conclude that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. (8) Because tensorflow-macos{sys_platform == 'darwin'}==2.10.0 depends on tensorflow-macos{sys_platform == 'darwin'}==2.10.0 and tensorflow-macos{sys_platform == 'darwin'}>=2.9.1,<=2.11.0 depends on protobuf>=3.9.2,<3.20, we can conclude that any of: tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 depend on protobuf>=3.9.2,<3.20. And because tensorflow-macos{sys_platform == 'darwin'}==2.8.0 depends on keras>=2.8.0rc0,<2.9 and flatbuffers>=1.12,<2, we can conclude that any of: flatbuffers<1.12 flatbuffers>=2 , any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , tensorflow-macos{sys_platform == 'darwin'}!=2.7.0, any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}==2.7.0 depends on flatbuffers>=1.12,<3.0 and tensorflow-cpu{sys_platform != 'darwin'}==2.16.2 depends on flatbuffers>=23.5.26, we can conclude that any of: keras<2.8.0rc0 keras>=2.9 , any of: protobuf<3.9.2 protobuf>=3.20 , tensorflow-cpu{sys_platform != 'darwin'}==2.16.2, any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because any of: tensorflow-cpu{sys_platform != 'darwin'}==2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>=2.17.0 depend on one of: protobuf>=3.20.3,<4.21.0 protobuf>4.21.0,<4.21.1 protobuf>4.21.1,<4.21.2 protobuf>4.21.2,<4.21.3 protobuf>4.21.3,<4.21.4 protobuf>4.21.4,<4.21.5 protobuf>4.21.5,<5.0.0.dev0 and keras>=3.0.0, we can conclude that tensorflow-cpu{sys_platform != 'darwin'}==2.16.2 and any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because we know from (8) that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.16.2 tensorflow-cpu{sys_platform != 'darwin'}>2.16.2,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible, we can conclude that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , any of: tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 are incompatible. And because only the following versions of tensorflow-macos{sys_platform == 'darwin'} are available: tensorflow-macos{sys_platform == 'darwin'}==2.5.0 tensorflow-macos{sys_platform == 'darwin'}==2.6.0 tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 tensorflow-macos{sys_platform == 'darwin'}==2.12.0 tensorflow-macos{sys_platform == 'darwin'}==2.13.0 tensorflow-macos{sys_platform == 'darwin'}==2.13.1 tensorflow-macos{sys_platform == 'darwin'}>=2.14.0 and tensorflow-macos{sys_platform == 'darwin'}>=2.13.0,<=2.13.1 depends on numpy>=1.22,<=1.24.3, we can conclude that any of: numpy<1.22 numpy>1.24.3 , any of: tensorflow-macos{sys_platform == 'darwin'}<2.5.0 tensorflow-macos{sys_platform == 'darwin'}>2.5.0,<2.6.0 tensorflow-macos{sys_platform == 'darwin'}>2.6.0,<2.12.0 tensorflow-macos{sys_platform == 'darwin'}>2.12.0 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , tensorflow-macos{sys_platform == 'darwin'}<2.14.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}==2.12.0 depends on numpy>=1.22,<1.24 and numpy>=1.19.2,<1.20.dev0, we can conclude that any of: numpy<1.19.2 numpy>=1.20.dev0,<1.22 numpy>1.24.3 , any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , tensorflow-macos{sys_platform == 'darwin'}<2.14.0 are incompatible. And because we know from (1) that colabfold[alphafold]==1.5.5, any of: numpy<1.26.0 numpy>1.26.0,<1.26.1 numpy>1.26.1,<1.26.2 numpy>1.26.2,<1.26.3 numpy>1.26.3,<2.0.0 numpy>2.0.0 , numpy{python_version >= '3.12'}!=1.26.4 are incompatible, we can conclude that colabfold[alphafold]==1.5.5, any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , tensorflow-macos{sys_platform == 'darwin'}<2.14.0 are incompatible. (9) Because only the following versions of tensorflow-macos{sys_platform == 'darwin'} are available: tensorflow-macos{sys_platform == 'darwin'}==2.5.0 tensorflow-macos{sys_platform == 'darwin'}==2.6.0 tensorflow-macos{sys_platform == 'darwin'}==2.7.0 tensorflow-macos{sys_platform == 'darwin'}==2.8.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.0 tensorflow-macos{sys_platform == 'darwin'}==2.9.1 tensorflow-macos{sys_platform == 'darwin'}==2.9.2 tensorflow-macos{sys_platform == 'darwin'}==2.10.0 tensorflow-macos{sys_platform == 'darwin'}==2.11.0 tensorflow-macos{sys_platform == 'darwin'}==2.12.0 tensorflow-macos{sys_platform == 'darwin'}==2.13.0 tensorflow-macos{sys_platform == 'darwin'}==2.13.1 tensorflow-macos{sys_platform == 'darwin'}>=2.14.0 and tensorflow-macos{sys_platform == 'darwin'}<=2.6.0 depends on h5py>=3.1.0,<3.2.dev0, we can conclude that any of: tensorflow-macos{sys_platform == 'darwin'}<2.6.0 tensorflow-macos{sys_platform == 'darwin'}>2.6.0,<2.7.0 tensorflow-macos{sys_platform == 'darwin'}>2.7.0,<2.8.0 tensorflow-macos{sys_platform == 'darwin'}>2.8.0,<2.9.0 tensorflow-macos{sys_platform == 'darwin'}>2.9.0,<2.9.1 tensorflow-macos{sys_platform == 'darwin'}>2.9.1,<2.9.2 tensorflow-macos{sys_platform == 'darwin'}>2.9.2,<2.10.0 tensorflow-macos{sys_platform == 'darwin'}>2.10.0,<2.11.0 tensorflow-macos{sys_platform == 'darwin'}>2.11.0,<2.12.0 tensorflow-macos{sys_platform == 'darwin'}>2.12.0,<2.13.0 tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1 tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0 depend on h5py>=3.1.0,<3.2.dev0. And because tensorflow-macos{sys_platform == 'darwin'}==2.6.0 depends on keras>=2.6,<3.dev0, we can conclude that any of: h5py<3.1.0 h5py>=3.2.dev0 , any of: keras<2.6 keras>=3.dev0 , any of: tensorflow-macos{sys_platform == 'darwin'}<2.7.0 tensorflow-macos{sys_platform == 'darwin'}>2.7.0,<2.8.0 tensorflow-macos{sys_platform == 'darwin'}>2.8.0,<2.9.0 tensorflow-macos{sys_platform == 'darwin'}>2.9.0,<2.9.1 tensorflow-macos{sys_platform == 'darwin'}>2.9.1,<2.9.2 tensorflow-macos{sys_platform == 'darwin'}>2.9.2,<2.10.0 tensorflow-macos{sys_platform == 'darwin'}>2.10.0,<2.11.0 tensorflow-macos{sys_platform == 'darwin'}>2.11.0,<2.12.0 tensorflow-macos{sys_platform == 'darwin'}>2.12.0,<2.13.0 tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1 tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}==2.7.0 depends on keras>=2.7.0rc0,<2.8 and keras>=2.8.0rc0,<2.9, we can conclude that any of: h5py<3.1.0 h5py>=3.2.dev0 , any of: keras<2.6 keras>=3.dev0 , any of: tensorflow-macos{sys_platform == 'darwin'}<2.9.0 tensorflow-macos{sys_platform == 'darwin'}>2.9.0,<2.9.1 tensorflow-macos{sys_platform == 'darwin'}>2.9.1,<2.9.2 tensorflow-macos{sys_platform == 'darwin'}>2.9.2,<2.10.0 tensorflow-macos{sys_platform == 'darwin'}>2.10.0,<2.11.0 tensorflow-macos{sys_platform == 'darwin'}>2.11.0,<2.12.0 tensorflow-macos{sys_platform == 'darwin'}>2.12.0,<2.13.0 tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1 tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}>=2.9.0,<=2.9.2 depends on keras>=2.9.0rc0,<2.10.0 and keras>=2.9.0rc0,<2.10.0, we can conclude that any of: h5py<3.1.0 h5py>=3.2.dev0 , any of: keras<2.6 keras>=3.dev0 , any of: tensorflow-macos{sys_platform == 'darwin'}<2.9.2 tensorflow-macos{sys_platform == 'darwin'}>2.9.2,<2.10.0 tensorflow-macos{sys_platform == 'darwin'}>2.10.0,<2.11.0 tensorflow-macos{sys_platform == 'darwin'}>2.11.0,<2.12.0 tensorflow-macos{sys_platform == 'darwin'}>2.12.0,<2.13.0 tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1 tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}==2.9.2 depends on keras>=2.9.0rc0,<2.10.0 and keras>=2.10.0,<2.11, we can conclude that any of: h5py<3.1.0 h5py>=3.2.dev0 , any of: keras<2.6 keras>=3.dev0 , any of: tensorflow-macos{sys_platform == 'darwin'}<2.11.0 tensorflow-macos{sys_platform == 'darwin'}>2.11.0,<2.12.0 tensorflow-macos{sys_platform == 'darwin'}>2.12.0,<2.13.0 tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1 tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}==2.11.0 depends on keras>=2.11.0,<2.12 and keras>=2.12.0,<2.13, we can conclude that any of: h5py<3.1.0 h5py>=3.2.dev0 , any of: keras<2.6 keras>=3.dev0 , any of: tensorflow-macos{sys_platform == 'darwin'}<2.13.0 tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1 tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0 are incompatible. And because tensorflow-macos{sys_platform == 'darwin'}>=2.13.0,<=2.13.1 depends on keras>=2.13.1,<2.14 and keras>=2.13.1,<2.14, we can conclude that any of: h5py<3.1.0 h5py>=3.2.dev0 , any of: keras<2.6 keras>=3.dev0 , tensorflow-macos{sys_platform == 'darwin'}<2.14.0 are incompatible. And because tensorflow-cpu{sys_platform != 'darwin'}==2.17.0 depends on keras>=3.2.0 and h5py>=3.10.0, we can conclude that tensorflow-cpu{sys_platform != 'darwin'}==2.17.0 and tensorflow-macos{sys_platform == 'darwin'}<2.14.0 are incompatible. And because we know from (9) that colabfold[alphafold]==1.5.5, any of: tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<2.17.0 tensorflow-cpu{sys_platform != 'darwin'}>2.17.0,<3.0.0 , tensorflow-macos{sys_platform == 'darwin'}<2.14.0 are incompatible, we can conclude that colabfold[alphafold]==1.5.5, tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<3.0.0, tensorflow-macos{sys_platform == 'darwin'}<2.14.0 are incompatible. And because colabfold==1.5.5 depends on tensorflow-cpu{sys_platform != 'darwin'}>=2.12.1,<3.0.0 and tensorflow-macos{sys_platform == 'darwin'}<2.14.0, we can conclude that colabfold==1.5.5 and colabfold[alphafold]==1.5.5 are incompatible. And because you require colabfold==1.5.5 and colabfold[alphafold]==1.5.5, we can conclude that the requirements are unsatisfiable. ```
zanieb commented 4 months ago

Related

zanieb commented 3 months ago

As of f2d67180388101ed24e70eb71e04fef1e3ddfc8e the error is

  × No solution found when resolving dependencies for split (python_full_version >= '3.12'):
  ╰─▶ Because colabfold[alphafold]==1.5.5 depends on jax>=0.4.20 and only the following versions of jax are available:
          jax<=0.4.20
          jax==0.4.21
          jax==0.4.22
          jax==0.4.23
          jax==0.4.24
          jax==0.4.25
          jax==0.4.26
          jax==0.4.27
          jax==0.4.28
          jax==0.4.29
          jax==0.4.30
          jax==0.4.31
      we can conclude that colabfold[alphafold]==1.5.5 depends on jax>=0.4.20.
      And because jax>=0.4.20 depends on numpy>=1.26.0, we can conclude that colabfold[alphafold]==1.5.5 depends on numpy>=1.26.0.
      (1)

      Because only the following versions of tensorflow-macos{sys_platform == 'darwin'} are available:
          tensorflow-macos{sys_platform == 'darwin'}==2.5.0
          tensorflow-macos{sys_platform == 'darwin'}==2.6.0
          tensorflow-macos{sys_platform == 'darwin'}==2.7.0
          tensorflow-macos{sys_platform == 'darwin'}==2.8.0
          tensorflow-macos{sys_platform == 'darwin'}==2.9.0
          tensorflow-macos{sys_platform == 'darwin'}==2.9.1
          tensorflow-macos{sys_platform == 'darwin'}==2.9.2
          tensorflow-macos{sys_platform == 'darwin'}==2.10.0
          tensorflow-macos{sys_platform == 'darwin'}==2.11.0
          tensorflow-macos{sys_platform == 'darwin'}==2.12.0
          tensorflow-macos{sys_platform == 'darwin'}==2.13.0
          tensorflow-macos{sys_platform == 'darwin'}==2.13.1
          tensorflow-macos{sys_platform == 'darwin'}>=2.14.0
      and tensorflow-macos{sys_platform == 'darwin'}==2.5.0 has no wheels with a matching Python ABI tag, we can conclude that any
      of:
          tensorflow-macos{sys_platform == 'darwin'}<2.6.0
          tensorflow-macos{sys_platform == 'darwin'}>2.6.0,<2.7.0
          tensorflow-macos{sys_platform == 'darwin'}>2.7.0,<2.8.0
          tensorflow-macos{sys_platform == 'darwin'}>2.8.0,<2.9.0
          tensorflow-macos{sys_platform == 'darwin'}>2.9.0,<2.9.1
          tensorflow-macos{sys_platform == 'darwin'}>2.9.1,<2.9.2
          tensorflow-macos{sys_platform == 'darwin'}>2.9.2,<2.10.0
          tensorflow-macos{sys_platform == 'darwin'}>2.10.0,<2.11.0
          tensorflow-macos{sys_platform == 'darwin'}>2.11.0,<2.12.0
          tensorflow-macos{sys_platform == 'darwin'}>2.12.0,<2.13.0
          tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1
          tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0
       are incompatible.
      And because tensorflow-macos{sys_platform == 'darwin'}==2.6.0 has no wheels with a matching Python ABI tag and
      tensorflow-macos{sys_platform == 'darwin'}==2.7.0 has no wheels with a matching Python ABI tag, we can conclude that any of:
          tensorflow-macos{sys_platform == 'darwin'}<2.8.0
          tensorflow-macos{sys_platform == 'darwin'}>2.8.0,<2.9.0
          tensorflow-macos{sys_platform == 'darwin'}>2.9.0,<2.9.1
          tensorflow-macos{sys_platform == 'darwin'}>2.9.1,<2.9.2
          tensorflow-macos{sys_platform == 'darwin'}>2.9.2,<2.10.0
          tensorflow-macos{sys_platform == 'darwin'}>2.10.0,<2.11.0
          tensorflow-macos{sys_platform == 'darwin'}>2.11.0,<2.12.0
          tensorflow-macos{sys_platform == 'darwin'}>2.12.0,<2.13.0
          tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1
          tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0
       are incompatible.
      And because tensorflow-macos{sys_platform == 'darwin'}==2.8.0 has no wheels with a matching Python ABI tag and
      tensorflow-macos{sys_platform == 'darwin'}==2.9.0 has no wheels with a matching Python ABI tag, we can conclude that any of:
          tensorflow-macos{sys_platform == 'darwin'}<2.9.1
          tensorflow-macos{sys_platform == 'darwin'}>2.9.1,<2.9.2
          tensorflow-macos{sys_platform == 'darwin'}>2.9.2,<2.10.0
          tensorflow-macos{sys_platform == 'darwin'}>2.10.0,<2.11.0
          tensorflow-macos{sys_platform == 'darwin'}>2.11.0,<2.12.0
          tensorflow-macos{sys_platform == 'darwin'}>2.12.0,<2.13.0
          tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1
          tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0
       are incompatible.
      And because tensorflow-macos{sys_platform == 'darwin'}==2.9.1 has no wheels with a matching Python ABI tag and
      tensorflow-macos{sys_platform == 'darwin'}==2.9.2 has no wheels with a matching Python ABI tag, we can conclude that any of:
          tensorflow-macos{sys_platform == 'darwin'}<2.10.0
          tensorflow-macos{sys_platform == 'darwin'}>2.10.0,<2.11.0
          tensorflow-macos{sys_platform == 'darwin'}>2.11.0,<2.12.0
          tensorflow-macos{sys_platform == 'darwin'}>2.12.0,<2.13.0
          tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1
          tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0
       are incompatible.
      And because tensorflow-macos{sys_platform == 'darwin'}==2.10.0 has no wheels with a matching Python ABI tag and
      tensorflow-macos{sys_platform == 'darwin'}==2.11.0 has no wheels with a matching Python ABI tag, we can conclude that any of:
          tensorflow-macos{sys_platform == 'darwin'}<2.12.0
          tensorflow-macos{sys_platform == 'darwin'}>2.12.0,<2.13.0
          tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1
          tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0
       are incompatible.
      And because tensorflow-macos{sys_platform == 'darwin'}>=2.13.0,<=2.13.1 depends on numpy>=1.22,<=1.24.3 and numpy>=1.22,<1.24,
      we can conclude that tensorflow-macos{sys_platform == 'darwin'}<2.14.0 depends on numpy>=1.22,<=1.24.3.
      And because we know from (1) that colabfold[alphafold]==1.5.5 depends on numpy>=1.26.0, we can conclude that
      colabfold[alphafold]==1.5.5 and tensorflow-macos{sys_platform == 'darwin'}<2.14.0 are incompatible.
      And because colabfold==1.5.5 depends on tensorflow-macos{sys_platform == 'darwin'}<2.14.0, we can conclude that
      colabfold==1.5.5 and colabfold[alphafold]==1.5.5 are incompatible.
      And because you require colabfold==1.5.5 and colabfold[alphafold]==1.5.5, we can conclude that your requirements are
      unsatisfiable.
dimbleby commented 2 months ago

error message aside: is it right that there is no solution? https://github.com/sokrypton/ColabFold/blob/v1.5.5/poetry.lock says there is.

charliermarsh commented 2 months ago

Not sure. Worth understanding, I'll take a look. The part of the error that I don't see in the dependency graph is: because jax>=0.4.20 depends on numpy>=1.26.0.

charliermarsh commented 2 months ago

Jax does have Requires-Dist: numpy >=1.26.0 ; python_version >= "3.12", so that's actually correct. I think there isn't a solution to this for Python 3.12, and that's why we fail. We try to solve for Python 3.11 and later.

charliermarsh commented 2 months ago

It would be fixed by https://github.com/astral-sh/uv/issues/6150, which we should probably do.

charliermarsh commented 2 months ago

I actually think the right fix is... have the user specify a range of Python versions to resolve for, separate from requires-python. I don't think users should be putting upper bounds on their requires-python. What you want to convey is: I don't care about having a valid resolution for Python 3.12 (as opposed to: my code only supports Python 3.11).

charliermarsh commented 2 months ago

So, I'd say it should actually be solved by https://github.com/astral-sh/uv/issues/4087.

dimbleby commented 2 months ago

Ah, the -p 3.11 bit is ignored then?

charliermarsh commented 2 months ago

Not quite -- it's not that it's ignored. It's that we solve for Python 3.11 as a minimum supported version.

charliermarsh commented 2 months ago

It's like: requires-python = ">=3.11".

dimbleby commented 2 months ago

That'd do it. Then there's definitely no solution, right from the start: https://github.com/sokrypton/ColabFold/blob/675f93a44eee6589a003164b047e7d4183073d1e/pyproject.toml#L22

konstin commented 2 months ago

The error with current main is:

  × No solution found when resolving dependencies for split (python_full_version >= '3.12'):
  ╰─▶ Because colabfold[alphafold]==1.5.5 depends on jax>=0.4.20 and jax>=0.4.20 depends on numpy>=1.26.0, we
      can conclude that colabfold[alphafold]==1.5.5 depends on numpy>=1.26.0. (1)

      Because only the following versions of tensorflow-macos{sys_platform == 'darwin'} are available:
          tensorflow-macos{sys_platform == 'darwin'}<=2.12.0
          tensorflow-macos{sys_platform == 'darwin'}>=2.13.0
      and tensorflow-macos{sys_platform == 'darwin'}<=2.11.0 has no wheels with a matching Python ABI tag, we
      can conclude that all of:
          tensorflow-macos{sys_platform == 'darwin'}<2.6.0
          tensorflow-macos{sys_platform == 'darwin'}>2.6.0,<2.7.0
          tensorflow-macos{sys_platform == 'darwin'}>2.7.0,<2.8.0
          tensorflow-macos{sys_platform == 'darwin'}>2.8.0,<2.9.0
          tensorflow-macos{sys_platform == 'darwin'}>2.9.0,<2.9.1
          tensorflow-macos{sys_platform == 'darwin'}>2.9.1,<2.9.2
          tensorflow-macos{sys_platform == 'darwin'}>2.9.2,<2.10.0
          tensorflow-macos{sys_platform == 'darwin'}>2.10.0,<2.11.0
          tensorflow-macos{sys_platform == 'darwin'}>2.11.0,<2.12.0
          tensorflow-macos{sys_platform == 'darwin'}>2.12.0,<2.13.0
          tensorflow-macos{sys_platform == 'darwin'}>2.13.0,<2.13.1
          tensorflow-macos{sys_platform == 'darwin'}>2.13.1,<2.14.0
       are incompatible.
      And because tensorflow-macos{sys_platform == 'darwin'}>=2.13.0,<=2.13.1 depends on numpy>=1.22,<=1.24.3
      and numpy>=1.22,<1.24, we can conclude that tensorflow-macos{sys_platform == 'darwin'}<2.14.0 depends
      on numpy>=1.22,<=1.24.3.
      And because we know from (1) that colabfold[alphafold]==1.5.5 depends on numpy>=1.26.0, we can conclude
      that colabfold[alphafold]==1.5.5 and tensorflow-macos{sys_platform == 'darwin'}<2.14.0 are incompatible.
      And because colabfold==1.5.5 depends on tensorflow-macos{sys_platform == 'darwin'}<2.14.0, we can conclude
      that colabfold==1.5.5 and colabfold[alphafold]==1.5.5 are incompatible.
      And because you require colabfold==1.5.5 and colabfold[alphafold]==1.5.5, we can conclude that your
      requirements are unsatisfiable.

This does not look bad anymore. We could collapse tensorflow-macos{sys_platform == 'darwin' some more and it's unhelpful that we're complaining about a lack of ABI when we're meaning specifically the python version in the build tag, but the error message overall is correct and explains the problem.

dimbleby commented 2 months ago

if uv is trying to solve for python >= "3.12", it would have been even simpler to notice directly that colabfold 1.5.5 requires python ">=3.9,<3.12"

konstin commented 2 months ago

We're ignoring upper bounds on requires-python in the resolver, we've made the experience that this leads to undesirable behavior (https://github.com/astral-sh/uv/issues/4022)

dimbleby commented 2 months ago

interesting choice! (I see I was in that issue already but either missed or forgot this outcome)

well I guess this issue is an example of that decision having downside

Even with the relatively improved error message: if the fact is that this project does not install on python 3.12 and had declared that it didn't install on python 3.12 - I guess users would rather have seen that message than this one.

(of course there is upside too, no doubt some projects will succeed in installing even though they declared that they would fail)