@@ -28,34 +28,34 @@ old build first. You often need to rebuild detectron2 after reinstalling PyTorch
2828
2929### Install Pre-Built Detectron2 (Linux only)
3030
31- Choose from this table to install [ v0.3 (Nov 2020 )] ( https://github.com/facebookresearch/detectron2/releases ) :
31+ Choose from this table to install [ v0.4 (Mar 2021 )] ( https://github.com/facebookresearch/detectron2/releases ) :
3232
33- <table class =" docutils " ><tbody ><th width =" 80 " > CUDA </th ><th valign =" bottom " align =" left " width =" 100 " >torch 1.7</th ><th valign =" bottom " align =" left " width =" 100 " >torch 1.6</th ><th valign =" bottom " align =" left " width =" 100 " >torch 1.5</th > <tr ><td align =" left " >11.0</td ><td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
33+ <table class =" docutils " ><tbody ><th width =" 80 " > CUDA </th ><th valign =" bottom " align =" left " width =" 100 " >torch 1.8</th ><th valign =" bottom " align =" left " width =" 100 " >torch 1.7</th ><th valign =" bottom " align =" left " width =" 100 " >torch 1.6</th > <tr ><td align =" left " >11.1</td ><td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
34+ https://dl.fbaipublicfiles.com/detectron2/wheels/cu111/torch1.8/index.html
35+ </code ></pre > </details > </td > <td align =" left " > </td > <td align =" left " > </td > </tr > <tr ><td align =" left " >11.0</td ><td align =" left " > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
3436 https://dl.fbaipublicfiles.com/detectron2/wheels/cu110/torch1.7/index.html
35- </code ></pre > </details > </td > <td align =" left " > </td > <td align =" left " > </td > </tr > <tr ><td align =" left " >10.2</td ><td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
37+ </code ></pre > </details > </td > <td align =" left " > </td > </tr > <tr ><td align =" left " >10.2</td ><td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
38+ https://dl.fbaipublicfiles.com/detectron2/wheels/cu102/torch1.8/index.html
39+ </code ></pre > </details > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
3640 https://dl.fbaipublicfiles.com/detectron2/wheels/cu102/torch1.7/index.html
3741</code ></pre > </details > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
3842 https://dl.fbaipublicfiles.com/detectron2/wheels/cu102/torch1.6/index.html
39- </code ></pre > </details > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
40- https://dl.fbaipublicfiles.com/detectron2/wheels/cu102/torch1.5/index.html
4143</code ></pre > </details > </td > </tr > <tr ><td align =" left " >10.1</td ><td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
44+ https://dl.fbaipublicfiles.com/detectron2/wheels/cu101/torch1.8/index.html
45+ </code ></pre > </details > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
4246 https://dl.fbaipublicfiles.com/detectron2/wheels/cu101/torch1.7/index.html
4347</code ></pre > </details > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
4448 https://dl.fbaipublicfiles.com/detectron2/wheels/cu101/torch1.6/index.html
45- </code ></pre > </details > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
46- https://dl.fbaipublicfiles.com/detectron2/wheels/cu101/torch1.5/index.html
47- </code ></pre > </details > </td > </tr > <tr ><td align =" left " >9.2</td ><td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
49+ </code ></pre > </details > </td > </tr > <tr ><td align =" left " >9.2</td ><td align =" left " > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
4850 https://dl.fbaipublicfiles.com/detectron2/wheels/cu92/torch1.7/index.html
4951</code ></pre > </details > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
5052 https://dl.fbaipublicfiles.com/detectron2/wheels/cu92/torch1.6/index.html
51- </code ></pre > </details > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
52- https://dl.fbaipublicfiles.com/detectron2/wheels/cu92/torch1.5/index.html
5353</code ></pre > </details > </td > </tr > <tr ><td align =" left " >cpu</td ><td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
54+ https://dl.fbaipublicfiles.com/detectron2/wheels/cpu/torch1.8/index.html
55+ </code ></pre > </details > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
5456 https://dl.fbaipublicfiles.com/detectron2/wheels/cpu/torch1.7/index.html
5557</code ></pre > </details > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
5658 https://dl.fbaipublicfiles.com/detectron2/wheels/cpu/torch1.6/index.html
57- </code ></pre > </details > </td > <td align =" left " ><details ><summary > install </summary ><pre ><code >python -m pip install detectron2 -f \
58- https://dl.fbaipublicfiles.com/detectron2/wheels/cpu/torch1.5/index.html
5959</code ></pre > </details > </td > </tr ></tbody ></table >
6060
6161
@@ -183,14 +183,16 @@ to match your local CUDA installation, or install a different version of CUDA to
183183C++ compilation errors from NVCC / NVRTC; "Unsupported gpu architecture"
184184</summary >
185185
186+ A few possibilities:
187+
1861881 . Local CUDA/NVCC version has to match the CUDA version of your PyTorch. Both can be found in ` python collect_env.py ` .
187189 When they are inconsistent, you need to either install a different build of PyTorch (or build by yourself)
188190 to match your local CUDA installation, or install a different version of CUDA to match PyTorch.
189191
1901922 . Local CUDA/NVCC version shall support the SM architecture (a.k.a. compute capability) of your GPU.
191193 The capability of your GPU can be found at [ developer.nvidia.com/cuda-gpus] ( https://developer.nvidia.com/cuda-gpus ) .
192194 The capability supported by NVCC is listed at [ here] ( https://gist.github.com/ax3l/9489132 ) .
193- If your NVCC version is too old, this can be workaround a lower capability. environment variable
195+ If your NVCC version is too old, this can be workaround by setting environment variable
194196 ` TORCH_CUDA_ARCH_LIST ` to a lower, supported capability.
195197
1961983 . The combination of NVCC and GCC you use is incompatible. You need to change one of their versions.
0 commit comments