A neural network model for quickly solving multiple-band light curves of contact binaries
This is a neural network model trained on theoretical light curves generated by PHOEBE, designed for analyzing contact binary light curves. The optimized architecture enables: 
Multi-band light curve processing - Simultaneous analysis of observations across different passbands
Comprehensive parameter determination - Accurate estimation of:
• Temperature ratio (T₂/T₁)
• Mass ratio (q)
• Orbital inclination (i)
• System potential (Ω)
• Contact degree (f)
• Component luminosities (L₁, L₂)
• Relative radii (r₁, r₂)
• Third light contribution (l₃)
• Spot parameters (when present)
The model has been compiled into a user-friendly Windows executable (.exe) file for convenient deployment. Users are advised to carefully review the following operational guidelines prior to execution.

Input Data Specifications:
Light curve data should be provided in CSV format and stored in the "Input Data" directory. Filenames should correspond to their respective passbands (naming conventions should follow "bandpass.txt" file). Each CSV file must contain two columns:  phase (column header: "Phase")
and normalized flux (column header: "Flux").

Parameter Configuration:
The "Par_input.csv" file defines the input physical parameters, free parameters should be marked as "-1", fixed parameters can be assigned their respective values. Note: If no spots are present, "relteff" must be fixed at 1, while all other spot-related parameters can be set to 0

MCMC Configuration:
A two-stage MCMC process is implemented. The parameter search ranges, walkers, and iterations for the MCMC are specified in "data.json".

Output parameters:
The results include:
"Par_output.csv" – Final determined physical parameters (with the last entry representing the goodness of fit).
"lc.png" – Visualization of the fitted light curves.
"LC_theoretical_**.csv" – Theoretical light curve data.
"MCMC_result.csv" – Parameter values for each iteration in the second MCMC stage.
"step1.h5" & "step2.h5" – Data files containing MCMC chain information for both stages.

Mass Ratio Constraint Handling
Since the mass ratio is restricted to the range [0, 1], the analysis must be performed twice: 1. Using primary minimum-to-phase conversion. 2. Using secondary minimum-to-phase conversion. The final result is selected based on the superior goodness of fit.