import os import numpy as np import torch from torch.utils.data import DataLoader from torch.optim.lr_scheduler import StepLR import torch.optim as optim import torch.nn as nn from CT_dataset import * from UNet3D import * from losses_and_metrics import * import utils import pandas as pd if __name__ == '__main__': torch.cuda.set_device(utils.get_avail_gpu() ) # assign which gpu will be used (only linux works) use_visdom = True train_list = './train_list.csv' val_list = './val_list.csv' model_path = './models' model_name = 'ALV_unet3d_patch64x64x64_1500_3labels_30samples' #remember to include the project title (e.g., ALV) checkpoint_name = 'latest_checkpoint.tar' num_classes = 3 num_channels = 1 num_epochs = 50 num_workers = 6 train_batch_size = 6
import os import numpy as np import torch import torch.nn as nn from meshsegnet import * import utils from easy_mesh_vtk import * import pandas as pd from losses_and_metrics_for_mesh import * from scipy.spatial import distance_matrix if __name__ == '__main__': torch.cuda.set_device(utils.get_avail_gpu()) # assign which gpu will be used (only linux works) model_path = './models' model_name = 'Mesh_Segementation_MeshSegNet_15_classes_60samples_best.tar' mesh_path = '' # need to define sample_filenames = ['Sample_0101_d.stl'] # need to define output_path = './outputs' if not os.path.exists(output_path): os.mkdir(output_path) num_classes = 15 num_channels = 15 # set model device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') model = MeshSegNet(num_classes=num_classes, num_channels=num_channels).to(device, dtype=torch.float)
import utils from easy_mesh_vtk.easy_mesh_vtk import * import pandas as pd from losses_and_metrics_for_mesh import * from scipy.spatial import distance_matrix import scipy.io as sio import matlab.engine import shutil import time #from sklearn.svm import SVC from thundersvm import SVC from sklearn.neighbors import KNeighborsClassifier if __name__ == '__main__': gpu_id = utils.get_avail_gpu() torch.cuda.set_device(gpu_id) # assign which gpu will be used (only linux works) upsampling_method = 'SVM' #upsampling_method = 'KNN' model_path = './models' model_name = 'Mesh_Segementation_MeshSegNet_15_classes_60samples_best.tar' mesh_path = './inputs' sample_filenames = ['upper_T0.stl', 'upper_T1_aligned.stl'] output_path = './outputs' if not os.path.exists(output_path): os.mkdir(output_path) num_classes = 15