VQNiche

Tokenization for Tissue Sections

Installation

This is tested with Python 3.10. To install the latest development version, first activate your Python Environment and install this package in editable mode as follows:

pip install -e .

Usage

The model training pipeline consists of the following steps:

1. Build the dataset (see vqniche.dataset.in_memory_dataset_blob.py).
2. Initialize the DataModule (see vqniche.dataloaders.in_memory_datamodule.py).
3. Initialize the Model (see vqniche.models.vqgraph.py).
4. Initialize the Logger, and Checkpoints, and Trainer (see vqniche.utils.initialize.py).
5. Train the Model.

Contact

For questions and issues, please reach out to am84@sanger.ac.uk.