Pythonfiles
This commit is contained in:
@@ -4,68 +4,60 @@ from typing import Dict, Tuple
|
||||
|
||||
@dataclass
|
||||
class StochastischesModellApriori:
|
||||
sigma_beob: Iterable[float] #σ der einzelnen Beobachtung
|
||||
group_beob: Iterable[int] #Gruppenzugehörigkeit jeder Beobachtung (Distanz, Richtung, GNSS, Nivellement,...,)
|
||||
sigma0_groups: Dict[int, float] = field(default_factory=dict) #σ0² für jede Gruppe
|
||||
|
||||
sigma_obs: Iterable[float] # σ_i
|
||||
group_ids: Iterable[int] # Gruppenzugehörigkeit der i-ten Beobachtung
|
||||
sigma0_sq_groups: Dict[int, float] = field(default_factory=dict)
|
||||
|
||||
def __post_init__(self):
|
||||
# In sympy-Objekte konvertieren
|
||||
self.sigma_obs = sp.Matrix(list(self.sigma_obs)) # Spaltenvektor
|
||||
self.group_ids = sp.Matrix(list(self.group_ids)) # Spaltenvektor
|
||||
self.sigma_beob = sp.Matrix(list(self.sigma_beob)) #Spaltenvektor
|
||||
self.group_beob = sp.Matrix(list(self.group_beob)) #Spaltenvektor
|
||||
|
||||
if self.sigma_obs.rows != self.group_ids.rows:
|
||||
if self.sigma_beob.rows != self.group_beob.rows:
|
||||
raise ValueError("sigma_obs und group_ids müssen gleich viele Einträge haben.")
|
||||
|
||||
# Fehlende Gruppen mit σ_0j^2 = 1.0 initialisieren
|
||||
unique_groups = sorted({int(g) for g in self.group_ids})
|
||||
unique_groups = sorted({int(g) for g in self.group_beob}) #jede Beobachtungsgruppe wird genau einmal berücksichtigt
|
||||
for g in unique_groups:
|
||||
if g not in self.sigma0_sq_groups:
|
||||
self.sigma0_sq_groups[g] = 1.0
|
||||
if g not in self.sigma0_groups: #Fehlende Gruppen mit σ_0j^2 = 1.0
|
||||
self.sigma0_groups[g] = 1.0
|
||||
|
||||
|
||||
@property
|
||||
|
||||
def n_obs(self) -> int:
|
||||
return int(self.sigma_obs.rows)
|
||||
def n_beob(self) -> int:
|
||||
return int(self.sigma_beob.rows)
|
||||
|
||||
|
||||
def build_Qll_P(self) -> Tuple[sp.Matrix, sp.Matrix]:
|
||||
|
||||
n = self.n_obs
|
||||
def aufstellen_Qll_P(self) -> Tuple[sp.Matrix, sp.Matrix]:
|
||||
n = self.n_beob
|
||||
Q_ll = sp.zeros(n, n)
|
||||
P = sp.zeros(n, n)
|
||||
|
||||
for i in range(n):
|
||||
sigma_i = self.sigma_obs[i, 0]
|
||||
g = int(self.group_ids[i, 0])
|
||||
sigma0_sq = self.sigma0_sq_groups[g]
|
||||
|
||||
q_ii = sigma_i**2
|
||||
Q_ll[i, i] = q_ii
|
||||
|
||||
P[i, i] = 1 / (sigma0_sq * q_ii)
|
||||
|
||||
sigma_i = self.sigma_beob[i, 0] #σ-Wert der i-ten Beobachtung holen
|
||||
g = int(self.group_beob[i, 0]) #Gruppenzugehörigkeit der Beobachtung bestimmen
|
||||
sigma0_sq = self.sigma0_groups[g] #Den Varianzfaktor der Gruppe holen
|
||||
q_ii = sigma_i**2 #σ² berechnen
|
||||
Q_ll[i, i] = q_ii #Diagonale
|
||||
P[i, i] = 1 / (sigma0_sq * q_ii) #durch VKS nicht mehr P=Qll^-1
|
||||
return Q_ll, P
|
||||
|
||||
|
||||
@staticmethod
|
||||
def _redundanz_pro_beobachtung(A: sp.Matrix, P: sp.Matrix) -> sp.Matrix:
|
||||
def redundanz_pro_beobachtung(A: sp.Matrix, P: sp.Matrix) -> sp.Matrix:
|
||||
n_beob = P.rows #Anzahl der Beobachtungen (Zeilen in P)
|
||||
n_param = A.cols #Anzahl der Unbekannten (Spalten in A)
|
||||
|
||||
n_obs = P.rows
|
||||
n_param = A.cols
|
||||
|
||||
# P^(1/2) aufbauen (diagonal, sqrt der Diagonale)
|
||||
sqrtP = sp.zeros(n_obs, n_obs)
|
||||
for i in range(n_obs):
|
||||
sqrtP = sp.zeros(n_beob, n_beob) #Wurzel von P (der Diagonale)
|
||||
for i in range(n_beob):
|
||||
sqrtP[i, i] = sp.sqrt(P[i, i])
|
||||
|
||||
A_tilde = sqrtP * A # Ã
|
||||
A_tilde = sqrtP * A
|
||||
|
||||
# M = (Ãᵀ Ã)^(-1)
|
||||
M = (A_tilde.T * A_tilde).inv()
|
||||
|
||||
r_vec = sp.zeros(n_obs, 1)
|
||||
r_vec = sp.zeros(n_beob, 1)
|
||||
|
||||
for i in range(n_obs):
|
||||
for i in range(n_beob):
|
||||
a_i = A_tilde.row(i) # 1 × n_param
|
||||
a_i_row = sp.Matrix([a_i]) # explizit 1×n-Matrix
|
||||
r_i = 1 - (a_i_row * M * a_i_row.T)[0, 0]
|
||||
@@ -74,29 +66,29 @@ class StochastischesModellApriori:
|
||||
return r_vec
|
||||
|
||||
|
||||
def varianzkomponenten_schaetzung(
|
||||
def varianzkomponentenschaetzung(
|
||||
self,
|
||||
v: sp.Matrix, # Residuenvektor (n × 1)
|
||||
A: sp.Matrix, # Designmatrix
|
||||
) -> Dict[int, float]:
|
||||
|
||||
if v.rows != self.n_obs:
|
||||
if v.rows != self.n_beob:
|
||||
raise ValueError("Länge von v passt nicht zur Anzahl Beobachtungen im Modell.")
|
||||
|
||||
# Aktuelle Gewichte
|
||||
Q_ll, P = self.build_Qll_P()
|
||||
Q_ll, P = self.aufstellen_Qll_P()
|
||||
|
||||
# Redundanzzahlen pro Beobachtung
|
||||
r_vec = self._redundanz_pro_beobachtung(A, P)
|
||||
r_vec = self.redundanz_pro_beobachtung(A, P)
|
||||
|
||||
new_sigma0_sq: Dict[int, float] = {}
|
||||
|
||||
# Für jede Gruppe j:
|
||||
unique_groups = sorted({int(g) for g in self.group_ids})
|
||||
unique_groups = sorted({int(g) for g in self.group_beob})
|
||||
|
||||
for g in unique_groups:
|
||||
# Indizes der Beobachtungen in dieser Gruppe
|
||||
idx = [i for i in range(self.n_obs) if int(self.group_ids[i, 0]) == g]
|
||||
idx = [i for i in range(self.n_beob) if int(self.group_beob[i, 0]) == g]
|
||||
if not idx:
|
||||
continue
|
||||
|
||||
@@ -113,10 +105,8 @@ class StochastischesModellApriori:
|
||||
|
||||
# als float rausgeben, kann man aber auch symbolisch lassen
|
||||
new_sigma0_sq[g] = float(sigma_hat_j_sq)
|
||||
|
||||
return new_sigma0_sq
|
||||
|
||||
def update_sigma0(self, new_sigma0_sq: Dict[int, float]) -> None:
|
||||
|
||||
for g, val in new_sigma0_sq.items():
|
||||
self.sigma0_sq_groups[int(g)] = float(val)
|
||||
self.sigma0_groups[int(g)] = float(val)
|
||||
Reference in New Issue
Block a user