Discrete Choice Model (0,1)

[n pourshian]

I have been trying to use Biogeme library to model the discrete Choice model between 2 options (Choice 0,1)
(GP in the dataset choice is 0 and EL   choice is 1)

The problem is where I had to produce GP lanes based on EL rows
for example, if I have 1 row EL I might have 100 GP rows. 

But the coefficients are not making sense.

Is there anybody here that could help me with this?

ASC_GP = Beta('ASC_GP', 0, None, None, 1)        # Estimated

ASC_ML = Beta('ASC_ML', 0, None, None, 0)        # Fixed to 0

beta_TTS = Beta('beta_TTS', 0, None, None, 0)    

beta_Price = Beta('beta_Price', 0, None, None, 0)

V = {

    1: ASC_ML + beta_TTS* (TTS) + beta_Price * Price_ML,  # ML

    0: ASC_GP  # GP (no cost, no time savings)

}

av = {0: 1, 1: 1}