xFormers 版本的选择

每次安装各种AI工具，最常遇到的问题就是Python / CUDA / PyTorch / xFormers因版本问题不能使用。这里简单介绍一下怎样去做到xFormers跟Python / CUDA / PyTorch的版本对齐。

对于xFormers，可以运行下面的命令：

python -m xformers.info

输出的结果是：

xFormers 0.0.28
memory_efficient_attention.ckF:                    unavailable
memory_efficient_attention.ckB:                    unavailable
memory_efficient_attention.ck_decoderF:            unavailable
memory_efficient_attention.ck_splitKF:             unavailable
memory_efficient_attention.cutlassF-pt:            available
memory_efficient_attention.cutlassB-pt:            available
memory_efficient_attention.fa2F@v2.5.6-pt:         available
memory_efficient_attention.fa2B@v2.5.6-pt:         available
memory_efficient_attention.fa3F@0.0.0:             unavailable
memory_efficient_attention.fa3B@0.0.0:             unavailable
memory_efficient_attention.triton_splitKF:         available
indexing.scaled_index_addF:                        available
indexing.scaled_index_addB:                        available
indexing.index_select:                             available
sequence_parallel_fused.write_values:              available
sequence_parallel_fused.wait_values:               available
sequence_parallel_fused.cuda_memset_32b_async:     available
sp24.sparse24_sparsify_both_ways:                  available
sp24.sparse24_apply:                               available
sp24.sparse24_apply_dense_output:                  available
sp24._sparse24_gemm:                               available
sp24._cslt_sparse_mm@0.0.0:                        available
swiglu.dual_gemm_silu:                             available
swiglu.gemm_fused_operand_sum:                     available
swiglu.fused.p.cpp:                                available
is_triton_available:                               True
pytorch.version:                                   2.4.1+cu121
pytorch.cuda:                                      available
gpu.compute_capability:                            8.6
gpu.name:                                          NVIDIA GeForce RTX 3060
dcgm_profiler:                                     unavailable
build.info:                                        available
build.cuda_version:                                1201
build.hip_version:                                 None
build.python_version:                              3.10.15
build.torch_version:                               2.4.1+cu121
build.env.TORCH_CUDA_ARCH_LIST:                    6.0+PTX 7.0 7.5 8.0+PTX 9.0a
build.env.PYTORCH_ROCM_ARCH:                       None
build.env.XFORMERS_BUILD_TYPE:                     Release
build.env.XFORMERS_ENABLE_DEBUG_ASSERTIONS:        None
build.env.NVCC_FLAGS:                              -allow-unsupported-compiler
build.env.XFORMERS_PACKAGE_FROM:                   wheel-v0.0.28
build.nvcc_version:                                12.1.66
source.privacy:                                    open source

如果不匹配，会出现错误信息，提示Python/CUDA/PyTorch哪个版本不匹配。可以根据下面网站找到应该安装哪个xFormers版本。

https://anaconda.org/xformers/xformers/files

比如我们现在是用的各个版本是：

•         python 3.10 (py310)
  •         CUDA 12.1 (cu12.1.0)
    •         PyTorch 2.4.1 (pyt2.4.1)

网站里面可以找到：

所以我们应该安装xformers==0.0.28.post1