steffen/Labeling-Primitives-With-Point2Net
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

New File Types, automatic detection and header parameters

Browse Source
This commit is contained in:
Si11ium 2019-07-30 15:11:11 +02:00
parent 4e38de9a5b
commit 0b9d03a25d
2 changed files with 26 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

32
dataset/shapenet.py
View File

@ -16,7 +16,9 @@ class CustomShapeNet(InMemoryDataset):
categories = {key: val for val, key in enumerate(['Box', 'Cone', 'Cylinder', 'Sphere'])} categories = {key: val for val, key in enumerate(['Box', 'Cone', 'Cylinder', 'Sphere'])}
def __init__(self, root, train=True, transform=None, pre_filter=None, pre_transform=None, **kwargs): def __init__(self, root, train=True, transform=None, pre_filter=None, pre_transform=None,
headers=True, **kwargs):
self.has_headers = headers
super(CustomShapeNet, self).__init__(root, transform, pre_transform, pre_filter) super(CustomShapeNet, self).__init__(root, transform, pre_transform, pre_filter)
path = self.processed_paths[0] if train else self.processed_paths[1] path = self.processed_paths[0] if train else self.processed_paths[1]
self.data, self.slices = torch.load(path) self.data, self.slices = torch.load(path)
@ -64,20 +66,26 @@ class CustomShapeNet(InMemoryDataset):
for pointcloud in tqdm(os.scandir(os.path.join(self.raw_dir, setting))): for pointcloud in tqdm(os.scandir(os.path.join(self.raw_dir, setting))):
if not os.path.isdir(pointcloud): if not os.path.isdir(pointcloud):
continue continue
for element in glob.glob(os.path.join(pointcloud.path, '*.dat')): paths = list()
if os.path.split(element)[-1] not in ['pc.dat']: for ext in ['dat', 'xyz']:
paths.extend(glob.glob(os.path.join(pointcloud.path, f'*.{ext}')))
for element in paths:
if all([x not in os.path.split(element)[-1] for x in ['pc.dat', 'pc.xyz']]):
# Assign training data to the data container # Assign training data to the data container
# Following the original logic; # Following the original logic;
# y should be the label; # y should be the label;
# pos should be the six dimensional vector describing: !its pos not points!! # pos should be the six dimensional vector describing: !its pos not points!!
# x,y,z,x_rot,y_rot,z_rot # x,y,z,x_rot,y_rot,z_rot
y_raw = os.path.splitext(element)[0].split('_')[-2]
# Get the y - Label
y_raw = next(i for i, v in enumerate(self.categories.keys()) if v.lower() in element.lower())
# y_raw = os.path.splitext(element)[0].split('_')[-2]
with open(element,'r') as f: with open(element,'r') as f:
headers = f.__next__() if self.has_headers:
# Check if there are no useable nodes in this file, header says 0. headers = f.__next__()
if not int(headers.rstrip().split(delimiter)[0]): # Check if there are no useable nodes in this file, header says 0.
continue if not int(headers.rstrip().split(delimiter)[0]):
# Get the y - Label continue
# Iterate over all rows # Iterate over all rows
src = [[float(x) if x not in ['-nan(ind)', 'nan(ind)'] else 0 src = [[float(x) if x not in ['-nan(ind)', 'nan(ind)'] else 0
@ -87,7 +95,7 @@ class CustomShapeNet(InMemoryDataset):
continue continue
# pos = points[:, :3] # pos = points[:, :3]
# norm = points[:, 3:] # norm = points[:, 3:]
y_all = [self.categories[y_raw]] * points.shape[0] y_all = [y_raw] * points.shape[0]
y = torch.as_tensor(y_all, dtype=torch.int) y = torch.as_tensor(y_all, dtype=torch.int)
# points = torch.as_tensor(points, dtype=torch.float) # points = torch.as_tensor(points, dtype=torch.float)
# norm = torch.as_tensor(norm, dtype=torch.float) # norm = torch.as_tensor(norm, dtype=torch.float)
@ -115,10 +123,10 @@ class ShapeNetPartSegDataset(Dataset):
Resample raw point cloud to fixed number of points. Resample raw point cloud to fixed number of points.
Map raw label from range [1, N] to [0, N-1]. Map raw label from range [1, N] to [0, N-1].
""" """
def __init__(self, root_dir, train=True, transform=None, npoints=1024): def __init__(self, root_dir, train=True, transform=None, npoints=1024, headers=True):
super(ShapeNetPartSegDataset, self).__init__() super(ShapeNetPartSegDataset, self).__init__()
self.npoints = npoints self.npoints = npoints
self.dataset = CustomShapeNet(root=root_dir, train=train, transform=transform) self.dataset = CustomShapeNet(root=root_dir, train=train, transform=transform, headers=headers)
def __getitem__(self, index): def __getitem__(self, index):
data = self.dataset[index] data = self.dataset[index]

8
main.py
View File

@ -7,6 +7,7 @@ https://github.com/dragonbook/pointnet2-pytorch/blob/master/main.py
import os import os
import sys import sys
from distutils.util import strtobool
import random import random
import numpy as np import numpy as np
import argparse import argparse
@ -35,12 +36,15 @@ parser.add_argument('--outf', type=str, default='checkpoint', help='output folde
parser.add_argument('--batch_size', type=int, default=8, help='input batch size') parser.add_argument('--batch_size', type=int, default=8, help='input batch size')
parser.add_argument('--test_per_batches', type=int, default=1000, help='run a test batch per training batches number') parser.add_argument('--test_per_batches', type=int, default=1000, help='run a test batch per training batches number')
parser.add_argument('--num_workers', type=int, default=4, help='number of data loading workers') parser.add_argument('--num_workers', type=int, default=4, help='number of data loading workers')
parser.add_argument('--headers', type=strtobool, default=True, help='if raw files come with headers')
opt = parser.parse_args() opt = parser.parse_args()
print(opt) print(opt)
# Random seed # Random seed
opt.manual_seed = 123 opt.manual_seed = 123
opt.headers = bool(opt.headers)
print('Random seed: ', opt.manual_seed) print('Random seed: ', opt.manual_seed)
random.seed(opt.manual_seed) random.seed(opt.manual_seed)
np.random.seed(opt.manual_seed) np.random.seed(opt.manual_seed)
@ -64,10 +68,10 @@ if __name__ == '__main__':
train_transform = GT.Compose([GT.NormalizeScale(), RotTransform, TransTransform]) train_transform = GT.Compose([GT.NormalizeScale(), RotTransform, TransTransform])
test_transform = GT.Compose([GT.NormalizeScale(), ]) test_transform = GT.Compose([GT.NormalizeScale(), ])
dataset = ShapeNetPartSegDataset(root_dir=opt.dataset, train=True, transform=train_transform, npoints=opt.npoints) dataset = ShapeNetPartSegDataset(root_dir=opt.dataset, train=True, transform=train_transform, npoints=opt.npoints, headers=opt.headers)
dataLoader = DataLoader(dataset, batch_size=opt.batch_size, shuffle=True, num_workers=opt.num_workers) dataLoader = DataLoader(dataset, batch_size=opt.batch_size, shuffle=True, num_workers=opt.num_workers)
test_dataset = ShapeNetPartSegDataset(root_dir=opt.dataset, train=False, transform=test_transform, npoints=opt.npoints) test_dataset = ShapeNetPartSegDataset(root_dir=opt.dataset, train=False, transform=test_transform, npoints=opt.npoints, headers=opt.headers)
test_dataLoader = DataLoader(test_dataset, batch_size=opt.batch_size, shuffle=True, num_workers=opt.num_workers) test_dataLoader = DataLoader(test_dataset, batch_size=opt.batch_size, shuffle=True, num_workers=opt.num_workers)
num_classes = dataset.num_classes() num_classes = dataset.num_classes()