# Contributors
-Caffe is developed by a core set of BVLC members and the open-source community.
+Caffe is developed by a core set of BAIR members and the open-source community.
We thank all of our [contributors](https://github.com/BVLC/caffe/graphs/contributors)!
[](LICENSE)
Caffe is a deep learning framework made with expression, speed, and modularity in mind.
-It is developed by the Berkeley Vision and Learning Center ([BVLC](http://bvlc.eecs.berkeley.edu)) and community contributors.
+It is developed by Berkeley AI Research ([BAIR](http://bair.berkeley.edu))/The Berkeley Vision and Learning Center (BVLC) and community contributors.
Check out the [project site](http://caffe.berkeleyvision.org) for all the details like
- [DIY Deep Learning for Vision with Caffe](https://docs.google.com/presentation/d/1UeKXVgRvvxg9OUdh_UiC5G71UMscNPlvArsWER41PsU/edit#slide=id.p)
- [Tutorial Documentation](http://caffe.berkeleyvision.org/tutorial/)
-- [BVLC reference models](http://caffe.berkeleyvision.org/model_zoo.html) and the [community model zoo](https://github.com/BVLC/caffe/wiki/Model-Zoo)
+- [BAIR reference models](http://caffe.berkeleyvision.org/model_zoo.html) and the [community model zoo](https://github.com/BVLC/caffe/wiki/Model-Zoo)
- [Installation instructions](http://caffe.berkeleyvision.org/installation.html)
and step-by-step examples.
## License and Citation
Caffe is released under the [BSD 2-Clause license](https://github.com/BVLC/caffe/blob/master/LICENSE).
-The BVLC reference models are released for unrestricted use.
+The BAIR/BVLC reference models are released for unrestricted use.
Please cite Caffe in your publications if it helps your research:
<header>
<h1 class="header"><a href="/">Caffe</a></h1>
<p class="header">
- Deep learning framework by the <a class="header name" href="http://bvlc.eecs.berkeley.edu/">BVLC</a>
+ Deep learning framework by <a class="header name" href="http://bair.berkeley.edu/">BAIR</a>
</p>
<p class="header">
Created by
# Development and Contributing
Caffe is developed with active participation of the community.<br>
-The [BVLC](http://bvlc.eecs.berkeley.edu/) brewers welcome all contributions!
+The [BAIR](http://bair.berkeley.edu/)/BVLC brewers welcome all contributions!
The exact details of contributions are recorded by versioning and cited in our [acknowledgements](http://caffe.berkeleyvision.org/#acknowledgements).
This method is impartial and always up-to-date.
The `master` branch receives all new development including community contributions.
We try to keep it in a reliable state, but it is the bleeding edge, and things do get broken every now and then.
-BVLC maintainers will periodically make releases by marking stable checkpoints as tags and maintenance branches. [Past releases](https://github.com/BVLC/caffe/releases) are catalogued online.
+BAIR maintainers will periodically make releases by marking stable checkpoints as tags and maintenance branches. [Past releases](https://github.com/BVLC/caffe/releases) are catalogued online.
#### Issues & Pull Request Protocol
# Caffe
Caffe is a deep learning framework made with expression, speed, and modularity in mind.
-It is developed by the Berkeley Vision and Learning Center ([BVLC](http://bvlc.eecs.berkeley.edu)) and by community contributors.
+It is developed by Berkeley AI Research ([BAIR](http://bair.berkeley.edu)) and by community contributors.
[Yangqing Jia](http://daggerfs.com) created the project during his PhD at UC Berkeley.
Caffe is released under the [BSD 2-Clause license](https://github.com/BVLC/caffe/blob/master/LICENSE).
- [Installation instructions](/installation.html)<br>
Tested on Ubuntu, Red Hat, OS X.
* [Model Zoo](/model_zoo.html)<br>
-BVLC suggests a standard distribution format for Caffe models, and provides trained models.
+BAIR suggests a standard distribution format for Caffe models, and provides trained models.
* [Developing & Contributing](/development.html)<br>
Guidelines for development and contributing to Caffe.
* [API Documentation](/doxygen/annotated.html)<br>
## Acknowledgements
-The BVLC Caffe developers would like to thank NVIDIA for GPU donation, A9 and Amazon Web Services for a research grant in support of Caffe development and reproducible research in deep learning, and BVLC PI [Trevor Darrell](http://www.eecs.berkeley.edu/~trevor/) for guidance.
+The BAIR Caffe developers would like to thank NVIDIA for GPU donation, A9 and Amazon Web Services for a research grant in support of Caffe development and reproducible research in deep learning, and BAIR PI [Trevor Darrell](http://www.eecs.berkeley.edu/~trevor/) for guidance.
-The BVLC members who have contributed to Caffe are (alphabetical by first name):
+The BAIR members who have contributed to Caffe are (alphabetical by first name):
[Eric Tzeng](https://github.com/erictzeng), [Evan Shelhamer](http://imaginarynumber.net/), [Jeff Donahue](http://jeffdonahue.com/), [Jon Long](https://github.com/longjon), [Ross Girshick](http://www.cs.berkeley.edu/~rbg/), [Sergey Karayev](http://sergeykarayev.com/), [Sergio Guadarrama](http://www.eecs.berkeley.edu/~sguada/), and [Yangqing Jia](http://daggerfs.com/).
The open-source community plays an important and growing role in Caffe's development.
We sincerely appreciate your interest and contributions!
If you'd like to contribute, please read the [developing & contributing](development.html) guide.
-Yangqing would like to give a personal thanks to the NVIDIA Academic program for providing GPUs, [Oriol Vinyals](http://www1.icsi.berkeley.edu/~vinyals/) for discussions along the journey, and BVLC PI [Trevor Darrell](http://www.eecs.berkeley.edu/~trevor/) for advice.
+Yangqing would like to give a personal thanks to the NVIDIA Academic program for providing GPUs, [Oriol Vinyals](http://www1.icsi.berkeley.edu/~vinyals/) for discussions along the journey, and BAIR PI [Trevor Darrell](http://www.eecs.berkeley.edu/~trevor/) for advice.
## Where to get trained models
-First of all, we bundle BVLC-trained models for unrestricted, out of the box use.
+First of all, we bundle BAIR-trained models for unrestricted, out of the box use.
<br>
-See the [BVLC model license](#bvlc-model-license) for details.
+See the [BAIR model license](#bair-model-license) for details.
Each one of these can be downloaded by running `scripts/download_model_binary.py <dirname>` where `<dirname>` is specified below:
-- **BVLC Reference CaffeNet** in `models/bvlc_reference_caffenet`: AlexNet trained on ILSVRC 2012, with a minor variation from the version as described in [ImageNet classification with deep convolutional neural networks](http://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks) by Krizhevsky et al. in NIPS 2012. (Trained by Jeff Donahue @jeffdonahue)
-- **BVLC AlexNet** in `models/bvlc_alexnet`: AlexNet trained on ILSVRC 2012, almost exactly as described in [ImageNet classification with deep convolutional neural networks](http://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks) by Krizhevsky et al. in NIPS 2012. (Trained by Evan Shelhamer @shelhamer)
-- **BVLC Reference R-CNN ILSVRC-2013** in `models/bvlc_reference_rcnn_ilsvrc13`: pure Caffe implementation of [R-CNN](https://github.com/rbgirshick/rcnn) as described by Girshick et al. in CVPR 2014. (Trained by Ross Girshick @rbgirshick)
-- **BVLC GoogLeNet** in `models/bvlc_googlenet`: GoogLeNet trained on ILSVRC 2012, almost exactly as described in [Going Deeper with Convolutions](http://arxiv.org/abs/1409.4842) by Szegedy et al. in ILSVRC 2014. (Trained by Sergio Guadarrama @sguada)
+- **BAIR Reference CaffeNet** in `models/bvlc_reference_caffenet`: AlexNet trained on ILSVRC 2012, with a minor variation from the version as described in [ImageNet classification with deep convolutional neural networks](http://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks) by Krizhevsky et al. in NIPS 2012. (Trained by Jeff Donahue @jeffdonahue)
+- **BAIR AlexNet** in `models/bvlc_alexnet`: AlexNet trained on ILSVRC 2012, almost exactly as described in [ImageNet classification with deep convolutional neural networks](http://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks) by Krizhevsky et al. in NIPS 2012. (Trained by Evan Shelhamer @shelhamer)
+- **BAIR Reference R-CNN ILSVRC-2013** in `models/bvlc_reference_rcnn_ilsvrc13`: pure Caffe implementation of [R-CNN](https://github.com/rbgirshick/rcnn) as described by Girshick et al. in CVPR 2014. (Trained by Ross Girshick @rbgirshick)
+- **BAIR GoogLeNet** in `models/bvlc_googlenet`: GoogLeNet trained on ILSVRC 2012, almost exactly as described in [Going Deeper with Convolutions](http://arxiv.org/abs/1409.4842) by Szegedy et al. in ILSVRC 2014. (Trained by Sergio Guadarrama @sguada)
**Community models** made by Caffe users are posted to a publicly editable [wiki page](https://github.com/BVLC/caffe/wiki/Model-Zoo).
These models are subject to conditions of their respective authors such as citation and license.
### Hosting trained models
It is up to the user where to host the `.caffemodel` file.
-We host our BVLC-provided models on our own server.
+We host our BAIR-provided models on our own server.
Dropbox also works fine (tip: make sure that `?dl=1` is appended to the end of the URL).
`scripts/download_model_binary.py <dirname>` downloads the `.caffemodel` from the URL specified in the `<dirname>/readme.md` frontmatter and confirms SHA1.
-## BVLC model license
+## BAIR model license
-The Caffe models bundled by the BVLC are released for unrestricted use.
+The Caffe models bundled by the BAIR are released for unrestricted use.
These models are trained on data from the [ImageNet project](http://www.image-net.org/) and training data includes internet photos that may be subject to copyright.
# Hardware Configuration Assumptions
The current implementation uses a tree reduction strategy. e.g. if there are 4 GPUs in the system, 0:1, 2:3 will exchange gradients, then 0:2 (top of the tree) will exchange gradients, 0 will calculate
-updated model, 0\-\>2, and then 0\-\>1, 2\-\>3.
+updated model, 0\-\>2, and then 0\-\>1, 2\-\>3.
For best performance, P2P DMA access between devices is needed. Without P2P access, for example crossing PCIe root complex, data is copied through host and effective exchange bandwidth is greatly reduced.
# Scaling Performance
-Performance is **heavily** dependent on the PCIe topology of the system, the configuration of the neural network you are training, and the speed of each of the layers. Systems like the DIGITS DevBox have an optimized PCIe topology (X99-E WS chipset). In general, scaling on 2 GPUs tends to be ~1.8X on average for networks like AlexNet, CaffeNet, VGG, GoogleNet. 4 GPUs begins to have falloff in scaling. Generally with "weak scaling" where the batchsize increases with the number of GPUs you will see 3.5x scaling or so. With "strong scaling", the system can become communication bound, especially with layer performance optimizations like those in [cuDNNv3](http://nvidia.com/cudnn), and you will likely see closer to mid 2.x scaling in performance. Networks that have heavy computation compared to the number of parameters tend to have the best scaling performance.
\ No newline at end of file
+Performance is **heavily** dependent on the PCIe topology of the system, the configuration of the neural network you are training, and the speed of each of the layers. Systems like the DIGITS DevBox have an optimized PCIe topology (X99-E WS chipset). In general, scaling on 2 GPUs tends to be ~1.8X on average for networks like AlexNet, CaffeNet, VGG, GoogleNet. 4 GPUs begins to have falloff in scaling. Generally with "weak scaling" where the batchsize increases with the number of GPUs you will see 3.5x scaling or so. With "strong scaling", the system can become communication bound, especially with layer performance optimizations like those in [cuDNNv3](http://nvidia.com/cudnn), and you will likely see closer to mid 2.x scaling in performance. Networks that have heavy computation compared to the number of parameters tend to have the best scaling performance.
For training, each time point is 20 iterations/minibatches of 256 images for 5,120 images total. For testing, a 50,000 image validation set is classified.
-**Acknowledgements**: BVLC members are very grateful to NVIDIA for providing several GPUs to conduct this research.
+**Acknowledgements**: BAIR members are very grateful to NVIDIA for providing several GPUs to conduct this research.
## NVIDIA K40
* Run for a certain number of iterations and give back control to Matlab
* Intermingle arbitrary Matlab code with gradient steps
-An ILSVRC image classification demo is in caffe/matlab/demo/classification_demo.m (you need to download BVLC CaffeNet from [Model Zoo](http://caffe.berkeleyvision.org/model_zoo.html) to run it).
+An ILSVRC image classification demo is in caffe/matlab/demo/classification_demo.m (you need to download BAIR CaffeNet from [Model Zoo](http://caffe.berkeleyvision.org/model_zoo.html) to run it).
### Build MatCaffe
MatCaffe is very similar to PyCaffe in usage.
-Examples below shows detailed usages and assumes you have downloaded BVLC CaffeNet from [Model Zoo](http://caffe.berkeleyvision.org/model_zoo.html) and started `matlab` from caffe root folder.
+Examples below shows detailed usages and assumes you have downloaded BAIR CaffeNet from [Model Zoo](http://caffe.berkeleyvision.org/model_zoo.html) and started `matlab` from caffe root folder.
model = './models/bvlc_reference_caffenet/deploy.prototxt';
weights = './models/bvlc_reference_caffenet/bvlc_reference_caffenet.caffemodel';
# Fine-tuning CaffeNet for Style Recognition on "Flickr Style" Data
Fine-tuning takes an already learned model, adapts the architecture, and resumes training from the already learned model weights.
-Let's fine-tune the BVLC-distributed CaffeNet model on a different dataset, [Flickr Style](http://sergeykarayev.com/files/1311.3715v3.pdf), to predict image style instead of object category.
+Let's fine-tune the BAIR-distributed CaffeNet model on a different dataset, [Flickr Style](http://sergeykarayev.com/files/1311.3715v3.pdf), to predict image style instead of object category.
## Explanation
---
-name: BVLC AlexNet Model
+name: BAIR/BVLC AlexNet Model
caffemodel: bvlc_alexnet.caffemodel
caffemodel_url: http://dl.caffe.berkeleyvision.org/bvlc_alexnet.caffemodel
license: unrestricted
---
-name: BVLC GoogleNet Model
+name: BAIR/BVLC GoogleNet Model
caffemodel: bvlc_googlenet.caffemodel
caffemodel_url: http://dl.caffe.berkeleyvision.org/bvlc_googlenet.caffemodel
license: unrestricted
---
-name: BVLC CaffeNet Model
+name: BAIR/BVLC CaffeNet Model
caffemodel: bvlc_reference_caffenet.caffemodel
caffemodel_url: http://dl.caffe.berkeleyvision.org/bvlc_reference_caffenet.caffemodel
license: unrestricted
---
-name: BVLC Reference RCNN ILSVRC13 Model
+name: BAIR/BVLC Reference RCNN ILSVRC13 Model
caffemodel: bvlc_reference_rcnn_ilsvrc13.caffemodel
caffemodel_url: http://dl.caffe.berkeleyvision.org/bvlc_reference_rcnn_ilsvrc13.caffemodel
license: unrestricted
projects / platform / upstream / caffe.git / commitdiff