Adaptive Milling

Adaptive milling plugins for fibsemOS. This package currently contains:

• AdaptivePolishing: a milling strategy that uses machine learning to determine when polishing is complete.
• AsymmetricFiducial: a milling pattern that adds some asymmetry to the default Fiducial pattern.

Installation

Adaptive Milling is available from PyPI and can be installed via the following steps:

1. Decide where you want to create your Python virtual environment and open a terminal in that location.
2. Follow the one of the sets of instructions below. We recommend using 'uv' as it handles installing the appropriate torch backend automatically.
3. Create a desktop shortcut for AutoLamella (fibsemOS GUI):

• Run fibsem-autolamella-ui from the virtual environment created during step 3.
• Tools → Create Desktop Shortcut...

In order to get AutoLamella set up for your system, please see the 'Getting started' documentation for fibsemOS.

Install using uv (recommended)

# Create a virtual environment (Python 3.11 to match AutoScript's environment)
uv venv --python 3.11

# Activate the virtual environment (Windows)
.venv\Scripts\activate

# Install including AutoLamella from fibsemOS with the appropriate PyTorch backend for your machine
uv pip install adaptive-milling --extra ui --torch-backend auto

To activate the virtual environment on Linux systems: source .venv/bin/activate

Install using miniforge

# Create a virtual environment (Python 3.11 to match AutoScript's environment)
conda create -p ./.venv python=3.11

# Activate the virtual environment
conda activate ./.venv

# Install PyTorch (select appropriate compute platform, see table below)
python -m pip install torch torchvision --index-url <Index URL>

# Install including AutoLamella from fibsemOS
python -m pip install -e adaptive-milling[ui]

If using CUDA, the version must be below or equal to the the system CUDA version, which can be checked with the command nvidia-smi.

Usage

Please see the user guide.

Testing

You can run the package tests using pytest:

uv run pytest

Issues

Please use the GitHub issue tracker to submit bugs or request features.

Contributions

If you would like to help contribute to project, please read our contribution guide and code of conduct.

License

Copyright Rosalind Franklin Institute, 2024.

Distributed under the terms of the Apache-2.0 license with "Commons Clause" License Condition v1.0, see the license for further details.