MacOS 12.2.1 Intel (x86 chip) with AMD GPU acceleration

1. This was tested on macOS Monterey (12.2.1)

2. Install Miniforge for Intel-based Macs. (See project page for more info.)

3. Create a new environment for pixplot:

conda create --name pixplot python=3.8

conda activate pixplot

4. Install the required packages for Apple's support of Tensorflow on the AMD chips:

SYSTEM_VERSION_COMPAT=0 python -m pip install tensorflow-macos tensorflow-metal

5. Clone pixplot:

git clone https://github.com/YaleDHLab/pix-plot.git

6. Adjust the install_requires section of setup.py to read as follows:

  install_requires=[
    'cmake',
    'Cython',
    'glob2',
    'h5py',
    'iiif-downloader',
    'Pillow',
    'pointgrid>=0.0.2',
    'python-dateutil>=2.8.0',
    'scikit-learn',
    'scipy',
    'six',
    'tqdm',
    'umap-learn',
    'yale-dhlab-rasterfairy>=1.0.3',
    'yale-dhlab-keras-preprocessing>=1.1.1',
    'matplotlib'
  ],

7. Install numba via conda, not pip:

conda install -c numba numba

Fix a few issues:

pip install numpy==1.20.0 pytz

8. Run python setup.py install

9. Optional steps past this point:

pip install image_datasets (a helper utility to easily grab pixplot-ready datasets)

10. Get a sample dataset:

python

import image_datasets

image_datasets.oslomini.download()

(now exit python via quit() or CTRL+D)

11. Test the installation:

time pixplot --images="datasets/oslomini/images/*.jpg" --metadata="datasets/oslomini/metadata/metadata.csv" --shuffle