3735025322

Build Type

github

Committed by GitHub

Commit Message

Merge 4ba3734bd into 3422c3a43

Pull Request Pull Request #3012: Add make tutorial to docs CI workflow

Run Details

22222 of 24227 relevant lines covered (91.72%)

2.23 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

38.24

/pyro/contrib/examples/util.py

# Copyright (c) 2017-2019 Uber Technologies, Inc.
# SPDX-License-Identifier: Apache-2.0

import os
import sys

import torchvision.datasets as datasets
from torch.utils.data import DataLoader
from torchvision import transforms


class MNIST(datasets.MNIST):
    mirrors = [
        "https://d2hg8soec8ck9v.cloudfront.net/datasets/mnist/"
    ] + datasets.MNIST.mirrors


def get_data_loader(
    dataset_name,
    data_dir,
    batch_size=1,
    dataset_transforms=None,
    is_training_set=True,
    shuffle=True,
):
    if not dataset_transforms:
        dataset_transforms = []
    trans = transforms.Compose([transforms.ToTensor()] + dataset_transforms)
    if dataset_name == "MNIST":
        dataset = MNIST
    else:
        dataset = getattr(datasets, dataset_name)
    print("downloading data")
    dset = dataset(root=data_dir, train=is_training_set, transform=trans, download=True)
    print("download complete.")
    return DataLoader(dset, batch_size=batch_size, shuffle=shuffle)


def print_and_log(logger, msg):
    # print and log a message (if a logger is present)
    print(msg)
    sys.stdout.flush()
    if logger is not None:
        logger.write("{}\n".format(msg))
        logger.flush()


def get_data_directory(filepath=None):
    if "CI" in os.environ:
        return os.path.expanduser("~/.data")
    return os.path.abspath(os.path.join(os.path.dirname(filepath), ".data"))


def _mkdir_p(dirname):
    if not os.path.exists(dirname):
        try:
            os.makedirs(dirname)
        except FileExistsError:
            pass

1	# Copyright (c) 2017-2019 Uber Technologies, Inc.
2	# SPDX-License-Identifier: Apache-2.0
3
4	import os	1✔
5	import sys	1✔
6
7	import torchvision.datasets as datasets	1✔
8	from torch.utils.data import DataLoader	1✔
9	from torchvision import transforms	1✔
10
11
12	class MNIST(datasets.MNIST):	1✔
13	mirrors = [	1✔
14	"https://d2hg8soec8ck9v.cloudfront.net/datasets/mnist/"
15	] + datasets.MNIST.mirrors
16
17
18	def get_data_loader(	1✔
19	dataset_name,
20	data_dir,
21	batch_size=1,
22	dataset_transforms=None,
23	is_training_set=True,
24	shuffle=True,
25	):
26	if not dataset_transforms:	×
27	dataset_transforms = []	×
28	trans = transforms.Compose([transforms.ToTensor()] + dataset_transforms)	×
29	if dataset_name == "MNIST":	×
30	dataset = MNIST	×
31	else:
32	dataset = getattr(datasets, dataset_name)	×
33	print("downloading data")	×
34	dset = dataset(root=data_dir, train=is_training_set, transform=trans, download=True)	×
35	print("download complete.")	×
36	return DataLoader(dset, batch_size=batch_size, shuffle=shuffle)	×
37
38
39	def print_and_log(logger, msg):	1✔
40	# print and log a message (if a logger is present)
41	print(msg)	×
42	sys.stdout.flush()	×
43	if logger is not None:	×
44	logger.write("{}\n".format(msg))	×
45	logger.flush()	×
46
47
48	def get_data_directory(filepath=None):	1✔
49	if "CI" in os.environ:	1✔
50	return os.path.expanduser("~/.data")	1✔
51	return os.path.abspath(os.path.join(os.path.dirname(filepath), ".data"))	×
52
53
54	def _mkdir_p(dirname):	1✔
55	if not os.path.exists(dirname):	×
56	try:	×
57	os.makedirs(dirname)	×
58	except FileExistsError:	×
59	pass	×

pyro-ppl / pyro / 3735025322

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous