如何安装keras-bert？ （PackagesNotFoundError：当前频道不提供以下软件包）

Question

我正尝试按照以下说明安装keras-bert：BERT from R。本教程介绍了如何使用R从Keras加载和训练BERT模型。

但是当在Anaconda提示符（Windows）中运行时：

conda install keras-bert

我得到以下错误：

Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Collecting package metadata (repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.

PackagesNotFoundError: The following packages are not available from current channels:

  - keras-bert

Current channels:

  - https://repo.anaconda.com/pkgs/main/win-64
  - https://repo.anaconda.com/pkgs/main/noarch
  - https://repo.anaconda.com/pkgs/r/win-64
  - https://repo.anaconda.com/pkgs/r/noarch
  - https://repo.anaconda.com/pkgs/msys2/win-64
  - https://repo.anaconda.com/pkgs/msys2/noarch
  - https://conda.anaconda.org/conda-forge/win-64
  - https://conda.anaconda.org/conda-forge/noarch

To search for alternate channels that may provide the conda package you're
looking for, navigate to

    https://anaconda.org

and use the search bar at the top of the page.

我还导航到https://anaconda.org，搜索keras-bert，但是没有找到用于此搜索的项目。

我也尝试过：

python3 -m pip install keras-bert

我没有得到任何输出：

(base) C:\Users\Standard>python3 -m pip install keras-bert

(base) C:\Users\Standard>

但是当我去检查是否用R命令安装了keras-bert时

reticulate::py_module_available('keras_bert')

我获得了：

[1] FALSE

Answer 1

从YAML在新环境中安装

由于这需要将PyPI软件包与Conda混合使用，因此the best practice recommendation for this将使用YAML文件创建专用环境。在这里，您可能需要附加的版本约束来实现与本教程兼容的设置，但是此YAML足以让我执行第一步：

bert_env.yaml

name: bert_env
channels:
  - defaults
dependencies:
  - numpy
  - keras
  - pip
  - pip:
    - keras-bert

运行

conda env create -f bert_env.yaml

创建一个名为 bert_env 的环境。

然后我可以在R会话中运行

> reticulate::use_condaenv("bert_env", required=TRUE)

> reticulate::py_config()
# python:         /Users/user/miniconda3/envs/bert_env/bin/python
# libpython:      /Users/user/miniconda3/envs/bert_env/lib/libpython3.6m.dylib
# pythonhome:     /Users/user/miniconda3/envs/bert_env:/Users/user/miniconda3/envs/bert_env
# version:        3.6.10 |Anaconda, Inc.| (default, May  7 2020, 23:06:31)  [GCC 4.2.1 Compatible Clang 4.0.1 (tags/RELEASE_401/final)]
# numpy:          /Users/user/miniconda3/envs/bert_env/lib/python3.6/site-packages/numpy
# numpy_version:  1.18.5

> reticulate::py_module_available("keras_bert")
# [1] TRUE

> tensorflow::tf_config()
# TensorFlow v2.0.0 ()
# Python v3.6 (~/miniconda3/envs/bert_env/bin/python)

这都是在 osx-64 平台上运行的，因此 win-64 所下拉的版本可能会有所不同。最后，这主要应该是tweaking the YAML的问题。