crop_size=(352, 352))
train_loader = torch.utils.data.DataLoader(dataset=train_dataset,
batch_size=batch_size,
shuffle=True,
num_workers=4)
val_dataset = DatasetKITTIVal(kitti_depth_path=kitti_depth_path)
val_loader = torch.utils.data.DataLoader(dataset=val_dataset,
batch_size=val_batch_size,
shuffle=False,
num_workers=1)
criterion = MaskedL2Gauss().cuda()
rmse_criterion = RMSE().cuda()
model = DepthCompletionNet().cuda()
model = torch.nn.DataParallel(model)
model.train()
optimizer = torch.optim.Adam(model.parameters(),
lr=learning_rate,
weight_decay=weight_decay)
optimizer.zero_grad()
train_losses = []
batch_train_losses = []
val_losses = []
train_rmses = []
batch_train_rmses = []
val_rmses = []
for i_iter, batch in enumerate(train_loader):
eval_dataset = DatasetKITTIVal(kitti_depth_path=kitti_depth_path)
eval_loader = torch.utils.data.DataLoader(dataset=eval_dataset,
batch_size=batch_size,
shuffle=False,
num_workers=4)
criterion = MaskedL2Gauss().cuda()
rmse_criterion = RMSE().cuda()
for model_i in model_is:
print("model_i: %d" % model_i)
restore_from = "/root/evaluating_bdl/depthCompletion/trained_models/%s_%d/checkpoint_40000.pth" % (
model_id, model_i)
model = DepthCompletionNet().cuda()
model = torch.nn.DataParallel(model)
model.load_state_dict(torch.load(restore_from))
model.eval()
for M in M_values:
M_float = float(M)
print("M: %d" % M)
for run in range(num_runs_per_M):
print("run: %d" % run)
batch_losses = []
batch_rmses = []
sigma_alea_values = np.array([])
sigma_epi_values = np.array([])