@petronny Yes, libibverbs was merged into rdma-core. Package updated.
Search Criteria
Package Details: caffe2-cuda 0.8.2_1.1.0-1
| Git Clone URL: | https://aur.archlinux.org/caffe2.git (read-only) |
|---|---|
| Package Base: | caffe2 |
| Description: | A new lightweight, modular, and scalable deep learning framework (with cuda support) |
| Upstream URL: | https://caffe2.ai/ |
| Keywords: | ai artificial cuda intelligence nvidia |
| Licenses: | |
| Conflicts: | |
| Provides: | |
| Submitter: | dbermond |
| Maintainer: | dbermond |
| Last Packager: | dbermond |
| Votes: | 0 |
| Popularity: | 0.000000 |
| First Submitted: | 2017-04-29 16:21 |
| Last Updated: | 2019-05-02 02:22 |
Dependencies (34)
- blas (openblas-lapack-git, openblas-git, blas-tmg, flexiblas, blas-git, openblas-lapack-static, openblas-lapack-openmp, openblas-lapack, atlas-lapack, openblas)
- cuda (cuda65)
- cudnn
- gflags (gflags-git, gflags-static)
- google-glog (glog-git, google-glog-static)
- lapack (openblas-lapack-git, lapack-tmg, flexiblas, lapack-git, openblas-lapack-static, openblas-lapack-openmp, openblas-lapack, atlas-lapack)
- nccl (nccl-git)
- protobuf (protobuf3-git, protobuf-bin, protobuf-static)
- python (python-dbg)
- python-numpy (python-numpy-openblas, python-numpy-mkl)
- python-protobuf
- python-yaml
- cmake (cmake-git) (make)
- cuda (cuda65) (make)
- cudnn (make)
- git (git-git) (make)
- gtest (googletest-git) (make)
- nccl (nccl-git) (make)
- snappy (snappy-git, libsnappy, snappy-static) (make)
- graphviz (optional)
- python-flask (python-flask12, python-flask-git) (optional)
- python-future (optional)
- python-hypothesis (optional)
- python-lmdb (optional)
- python-matplotlib (python-matplotlib-git) (optional)
- python-nvd3 (optional)
- python-pydot (optional)
- python-requests (optional)
- python-scikit-image (optional)
- python-scipy (python-scipy-mkl, python-scipy-openblas) (optional)
- python-setuptools (optional)
- python-six (optional)
- python-tornado (optional)
- python-yaml (optional)
Required by (0)
Sources (29)
- git+https://github.com/benjaminp/six
- git+https://github.com/eigenteam/eigen-git-mirror
- git+https://github.com/facebook/zstd
- git+https://github.com/facebookincubator/gloo
- git+https://github.com/google/benchmark
- git+https://github.com/google/gemmlowp
- git+https://github.com/google/googletest
- git+https://github.com/google/protobuf
- git+https://github.com/houseroad/foxi
- git+https://github.com/intel/ARM_NEON_2_x86_SSE
- git+https://github.com/intel/ideep
- git+https://github.com/Maratyszcza/cpuinfo
- git+https://github.com/Maratyszcza/FP16
- git+https://github.com/Maratyszcza/FXdiv
- git+https://github.com/Maratyszcza/NNPACK
- git+https://github.com/Maratyszcza/PeachPy
- git+https://github.com/Maratyszcza/psimd
- git+https://github.com/Maratyszcza/pthreadpool
- git+https://github.com/NVIDIA/nccl.git
- git+https://github.com/NVlabs/cub
- git+https://github.com/onnx/onnx
- git+https://github.com/onnx/onnx-tensorrt
- git+https://github.com/PeachPy/enum34
- git+https://github.com/pybind/pybind11
- git+https://github.com/pytorch/fbgemm
- git+https://github.com/pytorch/pytorch.git#tag=v1.1.0
- git+https://github.com/pytorch/QNNPACK
- git+https://github.com/Yangqing/ios-cmake
- sleef-zdevito
dbermond commented on 2018-10-25 03:40
petronny commented on 2018-10-22 05:46
libibverbs no longer exists in AUR now ... Should we use rdma-core?
dbermond commented on 2018-08-22 18:15
Important notice:
This package now provides the non-cuda version (as known as the 'cpu only' build).
If you want caffe2 with cuda support, use package caffe2-cuda. This new package naming schema will better reflect the package contents. This conforms to tensorflow package naming from the official repositories.
dbermond commented on 2018-06-03 15:57
@petronny Thank you for pointing this. Pytorch checksum updated.
petronny commented on 2018-06-03 14:29
Wrong hashsum for pytorch-0.4.0.tar.gz Please update it to f91c059710f802c91bed8207f2d461851b1bc2d44f7cd6e9aaa548392db9412f
dbermond commented on 2018-04-30 19:27
PyTorch released its first stable version containing Caffe2.
Package updated to caffe2 0.8.2 from pytorch stable 0.4.0. Note that this package does not contain pytorch, but caffe2 only.
dbermond commented on 2018-04-30 19:26
PyTorch released its first stable version containing Caffe2.
Package updated to caffe2 0.8.2 from pytorch stable 0.4.0. Note that this package does not contain pytorch, but caffe2 only.
Note: gcc5 is needed for building. Fortunately, gcc5 was moved from AUR to the [community] official repository. No need to compile gcc5 from AUR for the time being.
dbermond commented on 2018-04-23 18:01
I have backported a patch to enable building with the newest version of 'gcc' from the official repositories (currently gcc7). Package is now building fine with gcc7.
dbermond commented on 2018-04-23 17:55
I have added patches for building with cuda 9.1. Package is finally building fine again.
Important notes:
-
gcc5 from the AUR is currently needed for building. It does not build with cuda 9.1 + gcc6 (this is a known upstream issue). Be warned that gcc5 from the AUR takes a lot of time to compile.
-
The Caffe2 source code moved to PyTorch repository, but currently there is no stable version containing Caffe2 there. It is yet to be released by upstream.
-
The current available stable version of Caffe2 does not have python3 support. Python3 support currently is available on the git master branch only. If you want python3 support right now, please use caffe2-git or caffe2-cpu-git packages.
dbermond commented on 2018-04-12 15:16
@daquexian Yes, I know, but currently there is no stable release that supports python3 yet. The python3 support is present in git master only.
Unfortunately, caffe2-git (and also caffe2) currently does not compile because upstream has a nvcc issue with cuda 9.1. Details here: https://github.com/caffe2/caffe2/issues/1459 (and also here: https://github.com/caffe2/caffe2/issues/1636).
If you want python3 support for caffe2 right now, the only option is to use the package caffe2-cpu-git.
Pinned Comments
dbermond commented on 2018-08-22 18:15
Important notice:
This package now provides the non-cuda version (as known as the 'cpu only' build).
If you want caffe2 with cuda support, use package caffe2-cuda. This new package naming schema will better reflect the package contents. This conforms to tensorflow package naming from the official repositories.