name for name in os.listdir(rootDir)
if os.path.isdir(os.path.join(rootDir, name))
]
trainingData = videoDataset2stream(folderList=listOfFolders,
rootDir=rootDir,
rootDir_rgb=rootDir_rgb,
N_FRAME=N_FRAME)
dataloader = DataLoader(trainingData,
batch_size=BATCH_SIZE,
shuffle=True,
num_workers=1)
## Initializing r, theta
P, Pall = gridRing(N)
Drr = abs(P)
Drr = torch.from_numpy(Drr).float()
Dtheta = np.angle(P)
Dtheta = torch.from_numpy(Dtheta).float()
def loadModel(ckpt_file):
loadedcheckpoint = torch.load(ckpt_file)
#model.load_state_dict(loadedcheckpoint['state_dict'])
#optimizer.load_state_dict(loadedcheckpoint['optimizer'])
stateDict = loadedcheckpoint['state_dict']
# load parameters
Dtheta = stateDict['l1.theta']
Drr = stateDict['l1.rr']
K = np.matmul(np.matmul(P_old,H.T),np.linalg.inv(S)) # (4,4) (4,1) (1,1)
I = F # (4,4)
x_new = x_old + (K * y_err).T # (1,4) = (1,4) + ((4,1)(1,1)).T
# print('correction after obs: ',(K * y_err).T)
P_new = np.matmul((I - np.matmul(K, H)), P_old) # (4,4) = ((4,4) - (4,1)(1,4)) (4,4)
x_old = x_new # (1,4)
P_old = np.matmul(np.matmul(F,P_new),F.T) + Q
return x_old, P_old
Po,Pall = gridRing(4)
poles = [Pall[0], Pall[2]]
print(poles)
Drr = abs(Po)
Dtheta = np.angle(Po)
WVar = []
# generate dictionary
for i in range(0, N): # matrix 8
W1 = (Drr**i) * mt.cos(i * Dtheta)
W2 = (-Drr**i) * mt.cos(i * Dtheta)
W3 = (Drr**i) * mt.sin(i * Dtheta)
W4 = (-Drr**i) * mt.sin(i * Dtheta)
W = np.concatenate((W1, W2, W3, W4), 0)
W = np.expand_dims(W, axis=0)
