def simple_deterministic(cls, products):
lambdas = generate_n_equal_numbers_that_sum_one(len(products))
ros = []
for i in range(len(products)):
row = generate_n_equal_numbers_that_sum_one(len(products) - 1)
ros.append(row[:i] + [0] + row[i:])
return cls(products, lambdas, ros)
def simple_deterministic(cls, products, amount_classes):
gammas = generate_n_equal_numbers_that_sum_one(amount_classes)
multi_logit_models = [
MultinomialLogitModel.simple_deterministic(products)
for i in range(amount_classes)
]
return cls(products, gammas, multi_logit_models)
def simple_deterministic(cls, products):
l = generate_n_equal_numbers_that_sum_one(len(products))
u = [1.0 for _ in products]
v = generate_n_equal_numbers_that_sum_m(len(products), m=1.0)
w = [1.0 for _ in products]
z = generate_n_equal_numbers_that_sum_m(len(products), m=0.0)
return cls(products, l, u, v, w, z)
def simple_deterministic(cls, products):
# Hay que anclar el primer parámetro, sólo importan los valores relativos a este.
return cls(products, generate_n_equal_numbers_that_sum_one(len(products) - 1))
def simple_deterministic(cls, products, ranked_lists):
betas = generate_n_equal_numbers_that_sum_one(len(ranked_lists))[1:]
return cls(products, ranked_lists, betas)
def simple_deterministic(cls, products, ranked_lists):
betas = generate_n_equal_numbers_that_sum_one(len(ranked_lists))[1:]
choices = [1 for _ in range(len(ranked_lists))]
return cls(products, ranked_lists, betas, choices)
def simple_deterministic(cls, products):
mus = generate_n_equal_numbers_that_sum_one(len(products))
sigmas = generate_n_equal_numbers_that_sum_one(len(products))
return cls(products, mus, sigmas)
def simple_deterministic(cls, products, nests):
for nest in nests:
nest['lambda'] = 0.8
etas = generate_n_equal_numbers_that_sum_one(len(products))
return cls(products, nests, etas)\