conda安装库时报错failed with initial frozen solve. Retrying with flexible solve.

conda安装库时报错：

conda install keras
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.

解决方案：

conda config --set channel_priority flexible

然后就正常运行了（换了更加灵活的方式？？？）

conda install keras
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.
Collecting package metadata (repodata.json): done
Solving environment: done

## Package Plan ##

  environment location: E:\Apps\Anaconda3

  added / updated specs:
    - keras


The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    tensorflow-base-1.15.0     |mkl_py37h190a33d_0        36.2 MB
    ------------------------------------------------------------
                                           Total:        36.2 MB

The following NEW packages will be INSTALLED:

  keras              pkgs/main/win-64::keras-2.2.4-0
  keras-base         pkgs/main/win-64::keras-base-2.2.4-py37_0

The following packages will be DOWNGRADED:

  tensorflow                       2.0.0-mkl_py37he1bbcac_0 --> 1.15.0-mkl_py37h3789bd0_0
  tensorflow-base                  2.0.0-mkl_py37hd1d5974_0 --> 1.15.0-mkl_py37h190a33d_0
  tensorflow-estima~                     2.0.0-pyh2649769_0 --> 1.15.1-pyh2649769_0


Proceed ([y]/n)? y

---

看了一下安装过程，好像是keras基于底层是tensorflow1.5，呃。。。

然后就放弃了keras，重构了下代码，用tensorflow.keras代替了keras。大部分的keras在tensorflow.keras都有，有些不能直接对应的自己去网上查查就行了。