tensorflow-resnet-master
所属分类:人工智能/神经网络/深度学习
开发工具:Python
文件大小:653KB
下载次数:13
上传日期:2018-09-26 19:01:02
上 传 者:
123xy
说明: ResNet在2015年被提出,在ImageNet比赛classification任务上获得第一名,因为它“简单与实用”并存,之后很多方法都建立在ResNet50或者ResNet101的基础上完成的,检测,分割,识别等领域都纷纷使用ResNet,Alpha zero也使用了ResNet,所以可见ResNet确实很好用。
(ResNet was proposed in 2015 and won the first place in the classification task of ImageNet competition. Since it was "simple and practical", many methods were built on ResNet50 or ResNet101. ResNet was used in detection, segmentation, recognition and other fields.)
文件列表:
LICENSE (1076, 2018-03-08)
__init__.py (21, 2018-03-08)
config.py (2852, 2018-03-08)
convert.py (9786, 2018-03-08)
data (0, 2018-03-08)
data\ResNet-101-deploy.prototxt (65439, 2018-03-08)
data\ResNet-152-deploy.prototxt (98034, 2018-03-08)
data\ResNet-50-deploy.prototxt (32520, 2018-03-08)
data\ResNet_mean.binaryproto (602126, 2018-03-08)
data\cat.jpg (140391, 2018-03-08)
data\tensorflow-resnet-pretrained-20160509.tar.gz.torrent (37393, 2018-03-08)
forward.py (660, 2018-03-08)
image_processing.py (17702, 2018-03-08)
resnet.py (10621, 2018-03-08)
resnet_train.py (4730, 2018-03-08)
synset.py (38823, 2018-03-08)
train_cifar.py (12088, 2018-03-08)
train_imagenet.py (2860, 2018-03-08)
This code is *very old* and doesn't run on modern TensorFlow. Indeed, TF now
includes a resnet model in its core library:
https://github.com/tensorflow/tensorflow/blob/master/tensorflow/contrib/slim/python/slim/nets/resnet_v1.py
Also checkout my new ML project at http://propelml.org/
# ResNet in TensorFlow
Implemenation of [Deep Residual Learning for Image
Recognition](http://arxiv.org/abs/1512.03385). Includes a tool to use He et
al's published trained Caffe weights in TensorFlow.
MIT license. Contributions welcome.
## Goals
* Be able to use the pre-trained model's that [Kaiming He has provided for
Caffe](https://github.com/KaimingHe/deep-residual-networks). The `convert.py`
will convert the weights for use with TensorFlow.
* Implemented in the style of
[Inception](https://github.com/tensorflow/models/tree/master/inception/inception)
not using any classes and making heavy use of variable scope. It should be
easily usable in other models.
* Foundation to experiment with changes to ResNet like [stochastic
depth](https://arxiv.org/abs/1603.09382), [shared weights at each
scale](https://arxiv.org/abs/1604.03***0), and 1D convolutions for audio. (Not yet implemented.)
* ResNet is fully convolutional and the implementation should allow inputs to be any size.
* Be able to train out of the box on CIFAR-10, 100, and ImageNet. (Implementation incomplete)
## Pretrained Model
To convert the published Caffe pretrained model, run `convert.py`. However
Caffe is annoying to install so I'm providing a download of the output of
convert.py:
[tensorflow-resnet-pretrained-20160509.tar.gz.torrent](https://raw.githubusercontent.com/ry/tensorflow-resnet/master/data/tensorflow-resnet-pretrained-20160509.tar.gz.torrent) 4***M
## Notes
* This code depends on [TensorFlow git commit
cf7ce8](https://github.com/tensorflow/tensorflow/commit/cf7ce8a7879b6a7ba90441724ea3f8353917a515)
or later because ResNet needs 1x1 convolutions with stride 2. TF 0.8 is not new
enough.
* The `convert.py` script checks that activations are similiar to the caffe version
but it's not exactly the same. This is probably due to differences between how
TF and Caffe handle padding. Also preprocessing is done with color-channel means
instead of pixel-wise means.
近期下载者:
相关文件:
收藏者: