<!--[![travis](https://travis-ci.org/pfnet/chainercv.svg?branch=master)](https://travis-ci.org/pfnet/chainercv)-->
<!--[![pypi](https://img.shields.io/pypi/v/chainercv.svg)](https://pypi.python.org/pypi/chainercv)-->
# ChainerCV
ChainerCV is a collection of tools to train neural networks for computer vision tasks using [Chainer](https://github.com/pfnet/chainer).
You can find the documentation [here](http://chainercv.readthedocs.io/en/latest/).
This project is under heavy development, and API has been changing rapidly.
# Installation
```
pip install chainercv
```
### Requirements
+ [Chainer](https://github.com/pfnet/chainer) and its dependencies
+ Pillow
For additional features
+ Matplotlib
+ OpenCV
+ Scikit-Learn
Environments under Python 2.7.12 and 3.6.0 are tested.
# Features
## Transforms
ChainerCV supports functions commonly used to prepare image data before feeding to neural networks.
We expect users to use these functions together with a dataset object (e.g. `chainer.dataset.DatasetMixin`).
Many of the datasets prepared in ChainerCV are very thin wrappers around raw datasets in the filesystem, and
the transforms work best with such thin dataset classes.
The users can create a custom preprocessing pipeline by defining a function that describes
procedures to transform data.
Here is an example where the user rescales input image and data augments it by randomly rotation.
```python
from chainer.datasets import get_mnist
from chainercv.datasets import TransformDataset
from chainercv.transforms import random_rotate
dataset, _ = get_mnist(ndim=3)
def transform(in_data):
# in_data is the returned values of VOCSemanticSegmentationDataset.get_example
img, label = in_data
img -= 0.5 # rescale to [-0.5, 0.5]
img = random_rotate(img)
return img, label
dataset = TransformDataset(dataset, transform)
img, label = dataset[0]
```
As found in the example, `random_rotate` is one of the transforms ChainerCV supports. Like other transforms, this is just a
function that takes an array as input.
Also, `TransformDataset` is a new dataset class added in ChainerCV that overrides the underlying dataset's `__getitem__` by calling `transform` as post processing.
# Automatic Download
ChainerCV supports automatic download of datasets. It uses Chainer's default download scheme for automatic download.
All data downloaded by ChainerCV is saved under a directory `$CHAINER_DATASET_ROOT/pfnet/chainercv`.
The default value of `$CHAINER_DATASET_ROOT` is `~/.chainer/dataset/`.
没有合适的资源?快使用搜索试试~ 我知道了~
PyPI 官网下载 | chainercv-0.4.5.tar.gz
1.该资源内容由用户上传,如若侵权请联系客服进行举报
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
版权申诉
5星 · 超过95%的资源 1 下载量 52 浏览量
2022-01-09
21:35:33
上传
评论
收藏 95KB GZ 举报
温馨提示
共175个文件
pyc:103个
py:62个
txt:4个
资源来自pypi官网。 资源全名:chainercv-0.4.5.tar.gz
资源推荐
资源详情
资源评论
收起资源包目录
PyPI 官网下载 | chainercv-0.4.5.tar.gz (175个子文件)
setup.cfg 38B
MANIFEST.in 48B
README.md 2KB
cub_keypoint_dataset_test_image_ids.npy 16KB
PKG-INFO 272B
PKG-INFO 272B
semantic_segmentation_vis_report.py 10KB
detection_vis_report.py 10KB
googlenet.py 9KB
cub_keypoint_dataset.py 5KB
voc_detection_dataset.py 4KB
measure_k_recall.py 3KB
extension_utils.py 3KB
resize_contain.py 3KB
online_products_dataset.py 3KB
non_maximum_suppression.py 3KB
voc_semantic_segmentation_dataset.py 3KB
download.py 3KB
random_expand.py 3KB
resize.py 2KB
embed_images.py 2KB
vis_bbox.py 2KB
random_crop.py 2KB
test_utils.py 2KB
cub_label_dataset.py 2KB
vis_keypoint.py 2KB
cub_utils.py 2KB
center_crop.py 2KB
transform_dataset.py 2KB
pca_lighting.py 2KB
random_flip.py 2KB
eval_pck.py 1KB
ten_crop.py 1KB
voc_utils.py 1KB
flip_bbox.py 1KB
eval_semantic_segmentation.py 1KB
__init__.py 1KB
resize_bbox.py 1KB
translate_bbox.py 1KB
chw_to_pil_image.py 1KB
random_rotate.py 1KB
flip_keypoint.py 1KB
resize_keypoint.py 997B
dataset_utils.py 996B
scale.py 775B
setup.py 668B
flip.py 616B
__init__.py 553B
__init__.py 542B
image_utils.py 536B
__init__.py 380B
__init__.py 322B
__init__.py 233B
__init__.py 56B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
__init__.py 0B
semantic_segmentation_vis_report.pyc 11KB
googlenet.pyc 11KB
detection_vis_report.pyc 10KB
semantic_segmentation_vis_report.cpython-36.pyc 9KB
googlenet.cpython-36.pyc 9KB
detection_vis_report.cpython-36.pyc 8KB
voc_detection.cpython-36.pyc 6KB
cub_keypoint_dataset.pyc 6KB
online_products_dataset.cpython-36.pyc 5KB
voc_detection_dataset.pyc 5KB
online_products_dataset.pyc 4KB
cub_keypoint_dataset.cpython-36.pyc 4KB
measure_k_recall.pyc 4KB
voc_semantic_segmentation_dataset.cpython-36.pyc 4KB
voc_semantic_segmentation_dataset.pyc 4KB
test_utils.cpython-27-PYTEST.pyc 4KB
test_utils.pyc 4KB
resize_contain.pyc 3KB
cub_label_dataset.cpython-36.pyc 3KB
measure_k_recall.cpython-36.pyc 3KB
download.pyc 3KB
non_maximum_suppression.pyc 3KB
extension_utils.pyc 3KB
embed_images.pyc 3KB
test_utils.cpython-36-PYTEST.pyc 3KB
random_expand.pyc 3KB
cub_utils.pyc 3KB
cub_label_dataset.pyc 3KB
download.cpython-36.pyc 3KB
transform_dataset.pyc 3KB
embed_images.cpython-36.pyc 2KB
vis_keypoint.pyc 2KB
共 175 条
- 1
- 2
资源评论
- HYHjdk2023-11-17发现一个宝藏资源,资源有很高的参考价值,赶紧学起来~
挣扎的蓝藻
- 粉丝: 13w+
- 资源: 15万+
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功