scheduler/scheduler_config.py

import jax.numpy as jnp
import jax
import torch
from dataclasses import dataclass
import sympy
import sympy as sp
from sympy import Matrix, Symbol
import math
from sde_redefined_param import SDEDimension
@dataclass
class SDEParameterizedMaxNoiseBaseLineConfig:
    name = "Custom"
    variable = Symbol('t', nonnegative=True, real=True)

    drift_dimension = SDEDimension.SCALAR 
    diffusion_dimension = SDEDimension.SCALAR
    diffusion_matrix_dimension = SDEDimension.SCALAR 

    # TODO (KLAUS): HANDLE THE PARAMETERS BEING Ø
    drift_parameters = Matrix([sympy.symbols("f1", real=True)])
    diffusion_parameters = Matrix([sympy.symbols("sigma_max", real=True, nonzero=True)])
    
    drift = 0

    sigma_min = 0.002 
    sigma_max = sympy.Abs(diffusion_parameters[0]) #80
    diffusion = sigma_min * (sigma_max/sigma_min)**variable * sympy.sqrt(2 * sympy.Abs(sympy.log(sigma_max/sigma_min))) 

    # TODO (KLAUS) : in the SDE SAMPLING CHANGING Q impacts how we sample z ~ N(0, Q*(delta t))
    diffusion_matrix = 1 

    initial_variable_value = 0
    max_variable_value = 1 # math.inf
    min_sample_value = 0

    module = 'jax'

    drift_integral_form=False
    diffusion_integral_form=False
    diffusion_integral_decomposition = 'cholesky' # ldl

    non_symbolic_parameters = {'diffusion': torch.tensor([80.])}

    target = "epsilon" # x0