v1-interface.md

CryoLithe Legacy Interface (v1)

This document preserves the older inference workflow based on super.py and super-list.py.

Running the model

The script super.py is used to run the trained model on new projections. The script requires a YAML configuration file with the following fields:

• model_dir - path to the directory containing the trained model
• proj_file - path to the projection file
• angle_file - path to the angle file
• save_dir - path to the directory where the output will be saved
• save_name - name of the output volume
• device - CUDA device id to run the model on (integer, e.g. 0; older code also supports a list of ids, e.g. [0, 1])
• downsample_projections - whether to downsample the projections
• downsample_factor - factor by which to downsample the volume
• anti_alias - whether to apply anti-aliasing to the projections
• N3 - size of the volume along the z-axis
• batch_size - batch size to use when running the model

Run the script:

python3 super.py --config <path_to_config_file>

A sample YAML file is provided as docs/ribo80.yaml.

Running the model on the ribosome dataset

Download the ribosome dataset:

bash download_ribosome.sh

This places data in the data directory. The dataset includes:

• projections.mrcs - the projections
• angles.tlt - the angles

To run:

python3 super.py --config docs/ribo80.yaml

Using the wavelet model

python3 super.py --config docs/ribo80_wavelet.yaml

Running on a list of projections

Use super-list.py with docs/ribo80_list.yaml. You can set multiple values for:

• proj_file - list of projection paths
• angle_file - list of angle paths
• save_name - list of output names
• N3 - list of z-axis sizes

Run:

python3 super-list.py --config docs/ribo80_list.yaml

Downloading older models

bash download_old.sh