Installing DeepForest on Intel Mac (macOS x86_64)

[bw4sz]

Here is a quick workaround for older laptops needing to install source locally.

Installing DeepForest on Intel Mac (macOS x86_64)

PyTorch dropped macOS x86_64 wheel support after version 2.2.2. If you use uv for package management and see an error like:

error: Distribution torch can't be installed because it doesn't have a source distribution or wheel for the current platform

hint: You're on macOS (macosx_*_x86_64), but torch only has wheels for ... macosx_11_0_arm64 ...

Apply these two local-only changes to pyproject.toml (do not commit them) and re-run uv sync.

1. Add a [tool.uv] section

This tells the resolver it must find packages with x86_64 wheels:

[tool.uv]
required-environments = [
    "sys_platform == 'linux' and platform_machine == 'x86_64'",
    "sys_platform == 'linux' and platform_machine == 'aarch64'",
    "sys_platform == 'darwin' and platform_machine == 'arm64'",
    "sys_platform == 'darwin' and platform_machine == 'x86_64'",
    "sys_platform == 'win32' and platform_machine == 'AMD64'",
]

2. Pin numpy < 2.0

In the [project] dependencies list, change the numpy entry to:

"numpy<2.0",

3. Re-sync

uv sync --all-extras --dev

This will resolve torch to 2.2.2 (the last version with Intel Mac wheels) and a compatible numpy.

These are local-only workarounds for development on Intel Macs. Do not commit these changes.