deepspatial.core.DeepSpatial#
- class deepspatial.core.DeepSpatial[source]#
DeepSpatial: Reconstructing True 3D Spatial Omics at Single-Cell Resolution.
- __init__()[source]#
Initializes an empty DeepSpatial instance. Configurations and paths are injected during functional method calls.
Methods
__init__()Initializes an empty DeepSpatial instance.
build_model([patch_size, hidden_size, ...])Instantiates the GiT network architecture and Flow Matching logic.
fit([max_epochs, save_dir, accelerator, ...])Executes the training loop using PyTorch Lightning.
load_checkpoint(ckpt_path[, config_path, ...])Loads model weights and metadata for inference or resuming.
reconstruct_between_slices(adata0, adata1, ...)Generates a 3D volume segment between two specific AnnData slices.
reconstruct_full_volume(adata_list, thickness)High-level API to reconstruct the entire 3D volume from a list of slices.
setup_data(adata_list[, spatial_key, z_key, ...])Prepares the data pipeline and calculates physical normalization statistics.