Dataset required for CMM and GC-Net simulations

Instructions for Using the Dataset to Reproduce the Paper’s Results.Paper: Human and artificial visual systems share a computational principle for transforming binocular disparity into depth representationhttps://www.nature.com/articles/s42003-025-08474-1Dataset Contents.The dataset comprises pre-simulated data files necessary for reproducing the figures presented in the paper. The contents of the dataset and their corresponding roles are as follows:· BEM_canonical: Contains cross-correlation weights based on the binocular energy model (Supplementary Figure 4).· checkpoint: Includes the pretrained GC-Net model.· CMM: Stores cross-correlation and cross-matching responses to random dot stereograms (RDSs), used in Figure 3.· epoch_7_iter_22601: Outputs from GC-Net used in Figures 4 and 5.· MVPA: Provides decoding results for Figure 2 and Supplementary Figures 1 and 2.· rds: Contains the generated RDS stimuli used for simulations (general use).· S2N: fMRI percent signal change data used in Supplementary Figure 3.· VTC_extract_smoothed: Smoothed VTC data extracted from BrainVoyager (general use).· VTC_normalized: Normalized VTC data (general use).· VTC_stimID: Parameters describing the stimulus presentation corresponding to the VTC data (general use).· wavelet: Contains wavelet analysis data (Supplementary Figure 5).Python Code Repository.The Python code for generating the figures in the paper is available on GitHub and can be cloned from the following link:https://github.com/wundari/CMM_modelRepository Folder Structure Overview.The top-level structure of the cloned repository includes the following primary folders:· Codes/· Data/· Plots/Please refer to the file folder_tree_structure.txt for the detailed folder hierarchy and organization.Organizing Dataset Files.After cloning the repository, organize the dataset as follows:Step 1: Move the following folders into the CMM_model/Data/ directory:· BEM_canonical· CMM· MVPA· rds· S2N· VTC_extract_smoothed· VTC_normalized· VTC_stimID· waveletStep 2: Create the following directory path:CMM_model/Codes/Python/gcnet/results/sceneflow/monkaa/shift_1.5_median_wrt_leftThen move the following folders into that directory:· checkpoint· epoch_7_iter_22601For additional details and step-by-step instructions on running the Python scripts to reproduce the paper’s plots, please visit the GitHub repository: https://github.com/wundari/CMM_model