| from small one page howto to huge articles all in one place
Last additions:
May, 25th 2007:
April, 26th 2007:
Apr, 10th. 2007:
| 
. 
Details of sci-libs/caffe2:
Description: A deep learning framework
Homepage: https://pytorch.org/
available versions:
| releases | alpha | amd64 | arm | hppa | ia64 | mips | ppc | ppc64 | ppc macos | s390 | sh | sparc | x86 | USE-Flags | dependencies | ebuild warnings |
|---|
| caffe2-2.5.1-r7 |
- | ~ | - | - | - | - | - | - | - | - | - | - | - | cuda
distributed
fbgemm
flash
gloo
mkl
mpi
nnpack
+numpy
onednn
openblas
opencl
openmp
qnnpack
rocm
xnnpack
| show | WARNING: caffe2 is being built with its default CUDA compute capabilities: 3.5 and 7.0.
These may not be optimal for your GPU.
To configure caffe2 with the CUDA compute capability that is optimal for your GPU,
set TORCH_CUDA_ARCH_LIST in your make.conf, and re-emerge caffe2.
For example, to use CUDA capability 7.5 & 3.5, add: TORCH_CUDA_ARCH_LIST=7.5 3.5
For a Maxwell model GPU, an example value would be: TORCH_CUDA_ARCH_LIST=Maxwell
You can look up your GPU's CUDA compute capability at https://developer.nvidia.com/cuda-gpus
or by running /opt/cuda/extras/demo_suite/deviceQuery | grep 'CUDA Capability'
| | caffe2-2.4.1-r4 |
- | ~ | - | - | - | - | - | - | - | - | - | - | - | cuda
distributed
fbgemm
flash
gloo
mkl
mpi
nnpack
+numpy
onednn
openblas
opencl
openmp
qnnpack
rocm
xnnpack
| show | WARNING: caffe2 is being built with its default CUDA compute capabilities: 3.5 and 7.0.
These may not be optimal for your GPU.
To configure caffe2 with the CUDA compute capability that is optimal for your GPU,
set TORCH_CUDA_ARCH_LIST in your make.conf, and re-emerge caffe2.
For example, to use CUDA capability 7.5 & 3.5, add: TORCH_CUDA_ARCH_LIST=7.5 3.5
For a Maxwell model GPU, an example value would be: TORCH_CUDA_ARCH_LIST=Maxwell
You can look up your GPU's CUDA compute capability at https://developer.nvidia.com/cuda-gpus
or by running /opt/cuda/extras/demo_suite/deviceQuery | grep 'CUDA Capability'
| | caffe2-2.4.0-r1 |
- | ~ | - | - | - | - | - | - | - | - | - | - | - | cuda
distributed
fbgemm
flash
gloo
mkl
mpi
nnpack
+numpy
onednn
openblas
opencl
openmp
qnnpack
rocm
xnnpack
| show | WARNING: caffe2 is being built with its default CUDA compute capabilities: 3.5 and 7.0.
These may not be optimal for your GPU.
To configure caffe2 with the CUDA compute capability that is optimal for your GPU,
set TORCH_CUDA_ARCH_LIST in your make.conf, and re-emerge caffe2.
For example, to use CUDA capability 7.5 & 3.5, add: TORCH_CUDA_ARCH_LIST=7.5 3.5
For a Maxwell model GPU, an example value would be: TORCH_CUDA_ARCH_LIST=Maxwell
You can look up your GPU's CUDA compute capability at https://developer.nvidia.com/cuda-gpus
or by running /opt/cuda/extras/demo_suite/deviceQuery | grep 'CUDA Capability'
|
Legend: + stable~ testing- not availablesome ebuild warning depend on specific use-flags or architectures, all ebuild-warnings are shown.
| 
| 
| 
- Running on 
-
:
: 