tomodrgn.commands.train_nn

Train a FTPositionalDecoder neural network to reconstruct a 3D density map given 2D images from a tilt series with known pose and CTF parameters

Functions

add_args
decode_batch	Decode a batch of particles represented by multiple images from per-particle latent embeddings and corresponding lattice positions to evaluate
loss_function	Calculate generative loss between reconstructed and input images
main
save_checkpoint	Evaluate the model with the currently loaded weights; save the resulting consensus volume and model.
train_batch	Train a FTPositionalDecoder model on a batch of tilt series particle images.