为什么我不能安装这些特定要求？

Question

我正在尝试创建一个 conda 环境，但它有很多 anaconda 似乎无法找到的依赖项，总是在下面发送相同的错误消息：Collecting package metadata (current_repodata.json): done    Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source. Collecting package metadata (repodata.json): done Solving environment: failed        PackagesNotFoundError: The following packages are not available from current channels:      - opencv==3.4.2=py37h6fd60c2_1  - libopencv==3.4.2=h7c891bd_1  - pixman==0.38.0=h1de35cc_0  - gettext==0.19.8.1=h15daf44_3  - pyyaml==5.2=py37h1de35cc_0  - libcxxabi==4.0.1=hcfea43d_1  - mss==3.0.1=pypi_0  - cffi==1.13.2=py37h33e799b_0  - libprotobuf==3.11.2=hd9629dc_0  - graphite2==1.3.13=h2098e52_0  - pyscreenshot==0.6=pypi_0  - scipy==1.3.1=py37h1a1e112_0  - tensorflow-base==1.15.0=mkl_py37h032239d_0  - jasper==2.0.14=h636a363_1  - tornado==6.0.3=py37h1de35cc_0  - bzip2==1.0.8=h1de35cc_0  - c-ares==1.15.0=h1de35cc_1001  - numpy==1.17.4=py37hfd23ef0_0  - pycosat==0.6.3=py37h0b31af3_1002  - cairo==1.14.12=hc4e6be7_4  - zstd==1.3.7=h5bba6e5_0  - grpcio==1.16.1=py37h044775b_1  - pango==1.42.4=h060686c_0  - protobuf==3.11.2=py37h0a44026_0  - libvpx==1.7.0=h378b8a2_0  - graphviz==2.40.1=hefbbd9a_2  - tk==8.6.8=ha441bb4_0  - libpng==1.6.37=ha441bb4_0  - yaml==0.1.7=hc338f04_2  - sqlite==3.30.1=ha441bb4_0  - libcxx==4.0.1=hcfea43d_1  - harfbuzz==1.8.8=hb8d4a28_0  - pillow==6.2.1=py37hb68e598_0  - libiconv==1.15=hdd342a3_7  - pyobjc-core==5.2=py37_0  - easyprocess==0.2.8=pypi_0  - readline==7.0=h1de35cc_5  - xz==5.2.4=h1de35cc_4  - pcre==8.43=h0a44026_0  - glib==2.63.1=hd977a24_0  - libffi==3.2.1=h475c297_4  - ffmpeg==4.0=h01ea3c9_0  - libopenblas==0.3.6=hdc02c5d_2  - freetype==2.9.1=hb4e5f40_0我觉得这很奇怪，因为在附加所有主要频道后仍然出现此错误，我发现它们确实可用，例如 py-opencv-3.4.2-py37h7c891bd_1 可以在这里找到：https: //anaconda.org/anaconda /py-opencv/files?version=3.4.2。编辑：刚刚意识到这个项目只能在 Mac 上构建。有些依赖项与 Windows 不兼容，所以我想 RIP

Answer 1

py-opencv-3.4.2-py37h7c891bd_1是osx-64基于构建的，您正在尝试安装到 Windows 操作系统中。最好从您共享的链接中找到适用于 Windows 的相同版本。否则，只需使用没有内部版本号的版本。例如env.yml给你。

name: env-name
channels:
  - defaults
dependencies:
  - py-opencv==3.4.2