用于 ML 的 Databricks Runtime 5.5 LTS (EoS)

注意

对此 Databricks Runtime 版本的支持已结束。 有关终止支持日期,请参阅终止支持历史记录。 有关所有受支持的 Databricks Runtime 版本,请参阅 Databricks Runtime 发行说明版本和兼容性

Databricks 于 2019 年 7 月发布此版本。 支持已于 2021 年 7 月 27 日结束。 Databricks Runtime 5.5 ML 外延支持 (EoS) 将 5.5 ML 支持延长至 2021 年 12 月。 它使用 Ubuntu 18.04.5 LTS,而不是原始 Databricks Runtime 5.5 LTS 中使用的 Ubuntu 16.04.6 ML LTS 发行版(已弃用)。 Ubuntu 16.04.6 LTS 支持已于 2021 年 4 月 1 日终止。

用于机器学习的 Databricks Runtime 5.5 LTS 基于 Databricks Runtime 5.5 LTS (EoS) 为机器学习和数据科学提供随时可用的环境。 Databricks Runtime ML 包含许多常用的机器学习库,包括 TensorFlow、PyTorch、Keras 和 XGBoost。 它还支持使用 Horovod 进行分布式深度学习训练。

有关详细信息,包括有关如何创建 Databricks Runtime ML 群集的说明,请参阅 Databricks 上的 AI 和机器学习

新增功能

用于机器学习的 Databricks Runtime 5.5 LTS 是基于 Databricks Runtime 5.5 LTS 构建的。 若要了解 Databricks Runtime 5.5 LTS 中的新增功能,请参阅 Databricks Runtime 5.5 LTS (EoS) 发行说明。

除了库更新,用于机器学习的 Databricks Runtime 5.5 LTS 还引入了以下新功能:

  • 添加了 MLflow 1.0 Python 包

改进

  • 升级了机器学习库

    • TensorFlow 已从 1.12.0 升级到 1.13.1
    • PyTorch 已从 0.4.1 升级到 1.1.0
    • scikit-learn 已从 0.19.1 升级到 0.20.3
  • HorovodRunner 的单节点操作

    已启用 HorovodRunner,以便只在驱动程序节点上运行。 以前,若要使用 HorovodRunner,必须运行驱动程序和至少一个工作器节点。 而通过本次更改,你现可在单个节点(即多 GPU 节点)中分发训练,从而更有效地使用计算资源。

弃用

Hyperopt 库中,我们弃用了 hyperopt.SparkTrials 的以下属性:

  • SparkTrials.successful_trials_count
  • SparkTrials.failed_trials_count
  • SparkTrials.cancelled_trials_count
  • SparkTrials.total_trials_count

并将这些属性替换为以下函数:

  • SparkTrials.count_successful_trials()
  • SparkTrials.count_failed_trials()
  • SparkTrials.count_cancelled_trials()
  • SparkTrials.count_total_trials()

系统环境

用于机器学习的 Databricks Runtime 5.5 LTS 中的系统环境与 Databricks Runtime 5.5 不同,如下所示:

以下部分列出了用于机器学习的 Databricks Runtime 5.5 LTS 中包含的库,这些库与 Databricks Runtime 5.5 中包含的库不同。

顶层库

用于机器学习的 Databricks Runtime 5.5 LTS 包含以下顶层

Python 库

用于机器学习的 Databricks Runtime 5.5 LTS 使用 Conda 进行 Python 包管理。 因此,已安装的 Python 库相对于 Databricks Runtime 有很大区别。 以下部分介绍了用于机器学习的 Databricks Runtime 5.5 LTS 群集的 Conda 环境,这些群集使用 Python 2 或 3 以及支持 CPU 或 GPU 的计算机。

CPU 群集上的 Python 3

name: null
channels:
  - pytorch
  - defaults
dependencies:
  - _libgcc_mutex=0.1=main
  - _py-xgboost-mutex=2.0=cpu_0
  - _tflow_select=2.3.0=mkl
  - absl-py=0.7.1=py36_0
  - asn1crypto=0.24.0=py36_0
  - astor=0.7.1=py36_0
  - backcall=0.1.0=py36_0
  - backports=1.0=py_2
  - bcrypt=3.1.6=py36h7b6447c_0
  - blas=1.0=mkl
  - bleach=2.1.3=py36_0
  - boto=2.48.0=py36_1
  - boto3=1.7.62=py36h28b3542_1
  - botocore=1.10.62=py36h28b3542_0
  - ca-certificates=2018.03.07=0
  - certifi=2018.4.16=py36_0
  - cffi=1.11.5=py36he75722e_1
  - chardet=3.0.4=py36_1
  - click=7.0=py36_0
  - cloudpickle=0.8.0=py36_0
  - colorama=0.3.9=py36h489cec4_0
  - configparser=3.7.3=py36_1
  - cryptography=2.2.2=py36h14c3975_0
  - cycler=0.10.0=py36h93f1223_0
  - cython=0.28.2=py36h14c3975_0
  - decorator=4.3.0=py36_0
  - docutils=0.14=py36hb0f60f5_0
  - entrypoints=0.2.3=py36_2
  - et_xmlfile=1.0.1=py36hd6bccc3_0
  - flask=1.0.2=py36_1
  - freetype=2.8=hab7d2ae_1
  - gast=0.2.2=py36_0
  - gitdb2=2.0.5=py36_0
  - gitpython=2.1.11=py36_0
  - gmp=6.1.2=h6c8ec71_1
  - grpcio=1.12.1=py36hdbcaa40_0
  - gunicorn=19.9.0=py36_0
  - h5py=2.8.0=py36h989c5e5_3
  - hdf5=1.10.2=hba1933b_1
  - html5lib=1.0.1=py36_0
  - icu=58.2=h9c2bf20_1
  - idna=2.6=py36h82fb2a8_1
  - intel-openmp=2018.0.0=8
  - ipython=6.4.0=py36_1
  - ipython_genutils=0.2.0=py36_0
  - itsdangerous=0.24=py36_1
  - jdcal=1.4=py36_0
  - jedi=0.12.0=py36_1
  - jinja2=2.10=py36_0
  - jmespath=0.9.4=py_0
  - jpeg=9b=h024ee3a_2
  - jsonschema=2.6.0=py36_0
  - jupyter_client=5.2.3=py36_0
  - jupyter_core=4.4.0=py36_0
  - keras=2.2.4=0
  - keras-applications=1.0.8=py_0
  - keras-base=2.2.4=py36_0
  - keras-preprocessing=1.1.0=py_1
  - krb5=1.16.1=hc83ff2d_6
  - libedit=3.1.20170329=h6b74fdf_2
  - libffi=3.2.1=hd88cf55_4
  - libgcc-ng=7.3.0=hdf63c60_0
  - libgfortran-ng=7.2.0=hdf63c60_3
  - libpng=1.6.34=hb9fc6fc_0
  - libpq=10.4=h1ad7b7a_0
  - libprotobuf=3.8.0=hd408876_0
  - libsodium=1.0.16=h1bed415_0
  - libstdcxx-ng=7.3.0=hdf63c60_0
  - libtiff=4.0.9=he85c1e1_2
  - libxgboost=0.90=he6710b0_0
  - libxml2=2.9.8=h26e45fe_1
  - libxslt=1.1.32=h1312cb7_0
  - llvmlite=0.23.1=py36hdbcaa40_0
  - lxml=4.2.1=py36h23eabaa_0
  - mako=1.0.10=py_0
  - markdown=3.1.1=py36_0
  - markupsafe=1.0=py36h14c3975_1
  - mistune=0.8.3=py36h14c3975_1
  - mkl=2019.4=243
  - mkl_fft=1.0.12=py36ha843d7b_0
  - mkl_random=1.0.2=py36hd81dba3_0
  - mock=3.0.5=py36_0
  - msgpack-python=0.5.6=py36h6bb024c_1
  - nbconvert=5.3.1=py36_0
  - nbformat=4.4.0=py36h31c9010_0
  - ncurses=6.1=he6710b0_1
  - ninja=1.9.0=py36hfd86e86_0
  - numba=0.38.0=py36h637b7d7_0
  - numpy=1.16.2=py36h7e9f1db_0
  - numpy-base=1.16.2=py36hde5b4d6_0
  - olefile=0.45.1=py36_0
  - openpyxl=2.5.3=py36_0
  - openssl=1.0.2o=h14c3975_1
  - pandas=0.23.0=py36h637b7d7_0
  - pandocfilters=1.4.2=py36_1
  - paramiko=2.4.2=py36_0
  - parso=0.2.0=py36_0
  - pathlib2=2.3.2=py36_0
  - patsy=0.5.0=py36_0
  - pexpect=4.5.0=py36_0
  - pickleshare=0.7.4=py36_0
  - pillow=5.1.0=py36h3deb7b8_0
  - pip=10.0.1=py36_0
  - ply=3.11=py36_0
  - prompt_toolkit=1.0.15=py36h17d85b1_0
  - protobuf=3.8.0=py36he6710b0_0
  - psycopg2=2.7.5=py36hb7f436b_0
  - ptyprocess=0.5.2=py36h69acd42_0
  - py-xgboost=0.90=py36he6710b0_0
  - py-xgboost-cpu=0.90=py36_0
  - pyasn1=0.4.5=py_0
  - pycparser=2.18=py36_1
  - pygments=2.2.0=py36_0
  - pynacl=1.3.0=py36h7b6447c_0
  - pyopenssl=18.0.0=py36_0
  - pyparsing=2.2.0=py36_1
  - pysocks=1.6.8=py36_0
  - python=3.6.5=hc3d631a_2
  - python-dateutil=2.7.3=py36_0
  - python-editor=1.0.4=py_0
  - pytz=2018.4=py36_0
  - pyyaml=5.1=py36h7b6447c_0
  - pyzmq=17.0.0=py36h14c3975_3
  - readline=7.0=h7b6447c_5
  - requests=2.18.4=py36he2e5f8d_1
  - s3transfer=0.1.13=py36_0
  - scikit-learn=0.20.3=py36hd81dba3_0
  - scipy=1.1.0=py36h7c811a0_2
  - setuptools=39.1.0=py36_0
  - simplegeneric=0.8.1=py36_2
  - simplejson=3.16.0=py36h14c3975_0
  - singledispatch=3.4.0.3=py36_0
  - six=1.11.0=py36_1
  - smmap2=2.0.5=py36_0
  - sqlite=3.23.1=he433501_0
  - sqlparse=0.3.0=py_0
  - statsmodels=0.9.0=py36h035aef0_0
  - tabulate=0.8.3=py36_0
  - tensorboard=1.13.1=py36hf484d3e_0
  - tensorflow=1.13.1=mkl_py36h27d456a_0
  - tensorflow-base=1.13.1=mkl_py36h7ce6ba3_0
  - tensorflow-estimator=1.13.0=py_0
  - tensorflow-mkl=1.13.1=h4fcabd2_0
  - termcolor=1.1.0=py36_1
  - testpath=0.3.1=py36h8cadb63_0
  - tk=8.6.7=hc745277_3
  - tornado=5.0.2=py36h14c3975_0
  - traitlets=4.3.2=py36_0
  - urllib3=1.22=py36hbe7ace6_0
  - virtualenv=16.0.0=py36_0
  - wcwidth=0.1.7=py36hdf4376a_0
  - webencodings=0.5.1=py36_1
  - werkzeug=0.14.1=py36_0
  - wheel=0.31.1=py36_0
  - wrapt=1.11.1=py36h7b6447c_0
  - xz=5.2.4=h14c3975_4
  - yaml=0.1.7=had09818_2
  - zeromq=4.2.5=hf484d3e_1
  - zlib=1.2.11=h7b6447c_3
  - pytorch-cpu=1.1.0=py3.6_cpu_0
  - torchvision-cpu=0.3.0=py36_cuNone_1
  - pip:
    - databricks-cli==0.8.7
    - docker==4.0.2
    - fusepy==2.0.4
    - future==0.17.1
    - horovod==0.16.4
    - hyperopt==0.1.2.db6
    - kiwisolver==1.1.0
    - matplotlib==2.2.2
    - mleap==0.8.1
    - mlflow==1.0.0
    - msgpack==0.5.6
    - networkx==2.2
    - nose==1.3.7
    - nose-exclude==0.5.0
    - psutil==5.6.3
    - pyarrow==0.13.0
    - pymongo==3.8.0
    - querystring-parser==1.2.3
    - seaborn==0.8.1
    - tensorboardx==1.7
    - torchvision==0.3.0
    - tqdm==4.32.2
    - websocket-client==0.56.0
prefix: /databricks/python3

GPU 群集上的 Python 3

name: null
channels:
  - pytorch
  - Databricks
  - defaults
dependencies:
  - tensorflow=1.13.1.db1=gpu_py36h2903d8e_0
  - tensorflow-base=1.13.1.db1=gpu_py36he292aa2_0
  - tensorflow-gpu=1.13.1.db1=h0d30ee6_0
  - _libgcc_mutex=0.1=main
  - _py-xgboost-mutex=1.0=gpu_0
  - _tflow_select=2.1.0=gpu
  - absl-py=0.7.1=py36_0
  - asn1crypto=0.24.0=py36_0
  - astor=0.7.1=py36_0
  - backcall=0.1.0=py36_0
  - backports=1.0=py_2
  - bcrypt=3.1.6=py36h7b6447c_0
  - blas=1.0=mkl
  - bleach=2.1.3=py36_0
  - boto=2.48.0=py36_1
  - boto3=1.7.62=py36h28b3542_1
  - botocore=1.10.62=py36h28b3542_0
  - ca-certificates=2018.03.07=0
  - certifi=2018.4.16=py36_0
  - cffi=1.11.5=py36he75722e_1
  - chardet=3.0.4=py36_1
  - click=7.0=py36_0
  - cloudpickle=0.8.0=py36_0
  - colorama=0.3.9=py36h489cec4_0
  - configparser=3.7.3=py36_1
  - cryptography=2.2.2=py36h14c3975_0
  - cudnn=7.6.0=cuda10.0_0
  - cupti=10.0.130=0
  - cycler=0.10.0=py36_0
  - cython=0.28.2=py36h14c3975_0
  - decorator=4.3.0=py36_0
  - docutils=0.14=py36_0
  - entrypoints=0.2.3=py36_2
  - et_xmlfile=1.0.1=py36hd6bccc3_0
  - flask=1.0.2=py36_1
  - freetype=2.8=hab7d2ae_1
  - gast=0.2.2=py36_0
  - gitdb2=2.0.5=py36_0
  - gitpython=2.1.11=py36_0
  - gmp=6.1.2=h6c8ec71_1
  - grpcio=1.12.1=py36hdbcaa40_0
  - gunicorn=19.9.0=py36_0
  - h5py=2.8.0=py36h989c5e5_3
  - hdf5=1.10.2=hba1933b_1
  - html5lib=1.0.1=py36_0
  - icu=58.2=h9c2bf20_1
  - idna=2.6=py36h82fb2a8_1
  - intel-openmp=2018.0.0=8
  - ipython=6.4.0=py36_1
  - ipython_genutils=0.2.0=py36hb52b0d5_0
  - itsdangerous=0.24=py36_1
  - jdcal=1.4=py36_0
  - jedi=0.12.0=py36_1
  - jinja2=2.10=py36_0
  - jmespath=0.9.4=py_0
  - jpeg=9b=h024ee3a_2
  - jsonschema=2.6.0=py36_0
  - jupyter_client=5.2.3=py36_0
  - jupyter_core=4.4.0=py36_0
  - keras=2.2.4=0
  - keras-applications=1.0.8=py_0
  - keras-base=2.2.4=py36_0
  - keras-preprocessing=1.1.0=py_1
  - krb5=1.16.1=hc83ff2d_6
  - libedit=3.1.20170329=h6b74fdf_2
  - libffi=3.2.1=hd88cf55_4
  - libgcc-ng=7.3.0=hdf63c60_0
  - libgfortran-ng=7.2.0=hdf63c60_3
  - libpng=1.6.34=hb9fc6fc_0
  - libpq=10.4=h1ad7b7a_0
  - libprotobuf=3.8.0=hd408876_0
  - libsodium=1.0.16=h1bed415_0
  - libstdcxx-ng=7.3.0=hdf63c60_0
  - libtiff=4.0.9=he85c1e1_2
  - libxgboost=0.90=h688424c_0
  - libxml2=2.9.8=h26e45fe_1
  - libxslt=1.1.32=h1312cb7_0
  - llvmlite=0.23.1=py36hdbcaa40_0
  - lxml=4.2.1=py36h23eabaa_0
  - mako=1.0.10=py_0
  - markdown=3.1.1=py36_0
  - markupsafe=1.0=py36h14c3975_1
  - mistune=0.8.3=py36h14c3975_1
  - mkl=2019.4=243
  - mkl_fft=1.0.12=py36ha843d7b_0
  - mkl_random=1.0.2=py36hd81dba3_0
  - mock=3.0.5=py36_0
  - msgpack-python=0.5.6=py36h6bb024c_1
  - nbconvert=5.3.1=py36_0
  - nbformat=4.4.0=py36h31c9010_0
  - ncurses=6.1=he6710b0_1
  - ninja=1.9.0=py36hfd86e86_0
  - numba=0.38.0=py36h637b7d7_0
  - numpy=1.16.2=py36h7e9f1db_0
  - numpy-base=1.16.2=py36hde5b4d6_0
  - olefile=0.45.1=py36_0
  - openpyxl=2.5.3=py36_0
  - openssl=1.0.2o=h14c3975_1
  - pandas=0.23.0=py36h637b7d7_0
  - pandocfilters=1.4.2=py36_1
  - paramiko=2.4.2=py36_0
  - parso=0.2.0=py36_0
  - pathlib2=2.3.2=py36_0
  - patsy=0.5.0=py36_0
  - pexpect=4.5.0=py36_0
  - pickleshare=0.7.4=py36h63277f8_0
  - pillow=5.1.0=py36h3deb7b8_0
  - pip=10.0.1=py36_0
  - ply=3.11=py36_0
  - prompt_toolkit=1.0.15=py36_0
  - protobuf=3.8.0=py36he6710b0_0
  - psycopg2=2.7.5=py36hb7f436b_0
  - ptyprocess=0.5.2=py36h69acd42_0
  - py-xgboost=0.90=py36h688424c_0
  - py-xgboost-gpu=0.90=py36h28bbb66_0
  - pyasn1=0.4.5=py_0
  - pycparser=2.18=py36_1
  - pygments=2.2.0=py36_0
  - pynacl=1.3.0=py36h7b6447c_0
  - pyopenssl=18.0.0=py36_0
  - pyparsing=2.2.0=py36_1
  - pysocks=1.6.8=py36_0
  - python=3.6.5=hc3d631a_2
  - python-dateutil=2.7.3=py36_0
  - python-editor=1.0.4=py_0
  - pytz=2018.4=py36_0
  - pyyaml=5.1=py36h7b6447c_0
  - pyzmq=17.0.0=py36h14c3975_3
  - readline=7.0=h7b6447c_5
  - requests=2.18.4=py36he2e5f8d_1
  - s3transfer=0.1.13=py36_0
  - scikit-learn=0.20.3=py36hd81dba3_0
  - scipy=1.1.0=py36h7c811a0_2
  - setuptools=39.1.0=py36_0
  - simplegeneric=0.8.1=py36_2
  - simplejson=3.16.0=py36h14c3975_0
  - singledispatch=3.4.0.3=py36h7a266c3_0
  - six=1.11.0=py36_1
  - smmap2=2.0.5=py36_0
  - sqlite=3.23.1=he433501_0
  - sqlparse=0.3.0=py_0
  - statsmodels=0.9.0=py36h035aef0_0
  - tabulate=0.8.3=py36_0
  - tensorboard=1.13.1=py36hf484d3e_0
  - tensorflow-estimator=1.13.0=py_0
  - termcolor=1.1.0=py36_1
  - testpath=0.3.1=py36_0
  - tk=8.6.7=hc745277_3
  - tornado=5.0.2=py36h14c3975_0
  - traitlets=4.3.2=py36h674d592_0
  - urllib3=1.22=py36hbe7ace6_0
  - virtualenv=16.0.0=py36_0
  - wcwidth=0.1.7=py36hdf4376a_0
  - webencodings=0.5.1=py36_1
  - werkzeug=0.14.1=py36_0
  - wheel=0.31.1=py36_0
  - wrapt=1.11.1=py36h7b6447c_0
  - xz=5.2.4=h14c3975_4
  - yaml=0.1.7=had09818_2
  - zeromq=4.2.5=hf484d3e_1
  - zlib=1.2.11=h7b6447c_3
  - pytorch=1.1.0=py3.6_cuda10.0.130_cudnn7.5.1_0
  - torchvision=0.3.0=py36_cu10.0.130_1
  - pip:
    - databricks-cli==0.8.7
    - docker==4.0.2
    - fusepy==2.0.4
    - future==0.17.1
    - horovod==0.16.4
    - hyperopt==0.1.2.db6
    - kiwisolver==1.1.0
    - matplotlib==2.2.2
    - mleap==0.8.1
    - mlflow==1.0.0
    - msgpack==0.5.6
    - networkx==2.2
    - nose==1.3.7
    - nose-exclude==0.5.0
    - psutil==5.6.3
    - pyarrow==0.13.0
    - pymongo==3.8.0
    - querystring-parser==1.2.3
    - seaborn==0.8.1
    - tensorboardx==1.7
    - tqdm==4.32.2
    - websocket-client==0.56.0
prefix: /databricks/python3

CPU 群集上的 Python 2

name: null
channels:
  - pytorch
  - defaults
dependencies:
  - _libgcc_mutex=0.1=main
  - _py-xgboost-mutex=2.0=cpu_0
  - _tflow_select=2.3.0=mkl
  - absl-py=0.7.1=py27_0
  - asn1crypto=0.24.0=py27_0
  - astor=0.7.1=py27_0
  - backports=1.0=py_2
  - backports.shutil_get_terminal_size=1.0.0=py27_2
  - backports.weakref=1.0.post1=py_1
  - backports_abc=0.5=py_0
  - bcrypt=3.1.6=py27h7b6447c_0
  - blas=1.0=mkl
  - bleach=2.1.3=py27_0
  - boto=2.48.0=py27_1
  - boto3=1.7.62=py27h28b3542_1
  - botocore=1.10.62=py27h28b3542_0
  - ca-certificates=2018.03.07=0
  - certifi=2018.4.16=py27_0
  - cffi=1.11.5=py27he75722e_1
  - chardet=3.0.4=py27_1
  - click=7.0=py27_0
  - cloudpickle=0.8.0=py27_0
  - colorama=0.3.9=py27h5cde069_0
  - configparser=3.7.3=py27_1
  - cryptography=2.2.2=py27h14c3975_0
  - cycler=0.10.0=py27hc7354d3_0
  - cython=0.28.2=py27h14c3975_0
  - decorator=4.3.0=py27_0
  - docutils=0.14=py27_0
  - entrypoints=0.2.3=py27_2
  - enum34=1.1.6=py27_1
  - et_xmlfile=1.0.1=py27_0
  - flask=1.0.2=py27_1
  - freetype=2.8=hab7d2ae_1
  - funcsigs=1.0.2=py27_0
  - functools32=3.2.3.2=py27_1
  - future=0.17.1=py27_0
  - futures=3.2.0=py27_0
  - gast=0.2.2=py27_0
  - gitdb2=2.0.5=py27_0
  - gitpython=2.1.11=py27_0
  - gmp=6.1.2=h6c8ec71_1
  - grpcio=1.12.1=py27hdbcaa40_0
  - gunicorn=19.9.0=py27_0
  - h5py=2.8.0=py27h989c5e5_3
  - hdf5=1.10.2=hba1933b_1
  - html5lib=1.0.1=py27_0
  - icu=58.2=h9c2bf20_1
  - idna=2.6=py27h5722d68_1
  - intel-openmp=2018.0.0=8
  - ipaddress=1.0.22=py27_0
  - ipython=5.7.0=py27_0
  - ipython_genutils=0.2.0=py27_0
  - itsdangerous=0.24=py27_1
  - jdcal=1.4=py27_0
  - jinja2=2.10=py27_0
  - jmespath=0.9.4=py_0
  - jpeg=9b=h024ee3a_2
  - jsonschema=2.6.0=py27h7ed5aa4_0
  - jupyter_client=5.2.3=py27_0
  - jupyter_core=4.4.0=py27_0
  - keras=2.2.4=0
  - keras-applications=1.0.8=py_0
  - keras-base=2.2.4=py27_0
  - keras-preprocessing=1.1.0=py_1
  - krb5=1.16.1=hc83ff2d_6
  - libedit=3.1.20170329=h6b74fdf_2
  - libffi=3.2.1=hd88cf55_4
  - libgcc-ng=7.3.0=hdf63c60_0
  - libgfortran-ng=7.2.0=hdf63c60_3
  - libpng=1.6.34=hb9fc6fc_0
  - libpq=10.4=h1ad7b7a_0
  - libprotobuf=3.8.0=hd408876_0
  - libsodium=1.0.16=h1bed415_0
  - libstdcxx-ng=7.3.0=hdf63c60_0
  - libtiff=4.0.9=he85c1e1_2
  - libxgboost=0.90=he6710b0_0
  - libxml2=2.9.8=h26e45fe_1
  - libxslt=1.1.32=h1312cb7_0
  - linecache2=1.0.0=py27_0
  - llvmlite=0.23.1=py27hdbcaa40_0
  - lxml=4.2.1=py27h23eabaa_0
  - mako=1.0.10=py_0
  - markdown=3.1.1=py27_0
  - markupsafe=1.0=py27h14c3975_1
  - mistune=0.8.3=py27h14c3975_1
  - mkl=2019.4=243
  - mkl_fft=1.0.12=py27ha843d7b_0
  - mkl_random=1.0.2=py27hd81dba3_0
  - mock=3.0.5=py27_0
  - msgpack-python=0.5.6=py27h6bb024c_1
  - nbconvert=5.3.1=py27_0
  - nbformat=4.4.0=py27hed7f2b2_0
  - ncurses=6.1=he6710b0_1
  - ninja=1.9.0=py27hfd86e86_0
  - numba=0.38.0=py27h637b7d7_0
  - numpy=1.16.2=py27h7e9f1db_0
  - numpy-base=1.16.2=py27hde5b4d6_0
  - olefile=0.45.1=py27_0
  - openpyxl=2.5.3=py27_0
  - openssl=1.0.2o=h14c3975_1
  - pandas=0.23.0=py27h637b7d7_0
  - pandocfilters=1.4.2=py27_1
  - paramiko=2.4.2=py27_0
  - pathlib2=2.3.2=py27_0
  - patsy=0.5.0=py27_0
  - pexpect=4.5.0=py27_0
  - pickleshare=0.7.4=py27_0
  - pillow=5.1.0=py27h3deb7b8_0
  - pip=10.0.1=py27_0
  - ply=3.11=py27_0
  - prompt_toolkit=1.0.15=py27_0
  - protobuf=3.8.0=py27he6710b0_0
  - psycopg2=2.7.5=py27hb7f436b_0
  - ptyprocess=0.5.2=py27h4ccb14c_0
  - py-xgboost=0.90=py27he6710b0_0
  - py-xgboost-cpu=0.90=py27_0
  - pyasn1=0.4.5=py_0
  - pycparser=2.18=py27_1
  - pygments=2.2.0=py27_0
  - pynacl=1.3.0=py27h7b6447c_0
  - pyopenssl=18.0.0=py27_0
  - pyparsing=2.2.0=py27_1
  - pysocks=1.6.8=py27_0
  - python=2.7.15=h1571d57_0
  - python-dateutil=2.7.3=py27_0
  - python-editor=1.0.4=py_0
  - pytz=2018.4=py27_0
  - pyyaml=5.1=py27h7b6447c_0
  - pyzmq=17.0.0=py27h14c3975_3
  - readline=7.0=h7b6447c_5
  - requests=2.18.4=py27hc5b0589_1
  - s3transfer=0.1.13=py27_0
  - scandir=1.7=py27h14c3975_0
  - scikit-learn=0.20.3=py27hd81dba3_0
  - scipy=1.1.0=py27h7c811a0_2
  - setuptools=39.1.0=py27_0
  - simplegeneric=0.8.1=py27_2
  - simplejson=3.16.0=py27h14c3975_0
  - singledispatch=3.4.0.3=py27_0
  - six=1.11.0=py27_1
  - smmap2=2.0.5=py27_0
  - sqlite=3.23.1=he433501_0
  - sqlparse=0.3.0=py_0
  - statsmodels=0.9.0=py27h035aef0_0
  - tabulate=0.8.3=py27_0
  - tensorboard=1.13.1=py27hf484d3e_0
  - tensorflow=1.13.1=mkl_py27h74ee40f_0
  - tensorflow-base=1.13.1=mkl_py27h7ce6ba3_0
  - tensorflow-estimator=1.13.0=py_0
  - tensorflow-mkl=1.13.1=h4fcabd2_0
  - termcolor=1.1.0=py27_1
  - testpath=0.3.1=py27hc38d2c4_0
  - tk=8.6.7=hc745277_3
  - tornado=5.0.2=py27h14c3975_0
  - traceback2=1.4.0=py27_0
  - traitlets=4.3.2=py27_0
  - unittest2=1.1.0=py27_0
  - urllib3=1.22=py27ha55213b_0
  - virtualenv=16.0.0=py27_0
  - wcwidth=0.1.7=py27h9e3e1ab_0
  - webencodings=0.5.1=py27_1
  - werkzeug=0.14.1=py27_0
  - wheel=0.31.1=py27_0
  - wrapt=1.11.1=py27h7b6447c_0
  - xz=5.2.4=h14c3975_4
  - yaml=0.1.7=had09818_2
  - zeromq=4.2.5=hf484d3e_1
  - zlib=1.2.11=h7b6447c_3
  - pytorch-cpu=1.1.0=py2.7_cpu_0
  - torchvision-cpu=0.3.0=py27_cuNone_1
  - pip:
    - backports.functools-lru-cache==1.5
    - backports.ssl-match-hostname==3.7.0.1
    - databricks-cli==0.8.7
    - docker==4.0.2
    - fusepy==2.0.4
    - horovod==0.16.4
    - hyperopt==0.1.2.db6
    - kiwisolver==1.1.0
    - matplotlib==2.2.2
    - mleap==0.8.1
    - mlflow==1.0.0
    - msgpack==0.5.6
    - networkx==2.2
    - nose==1.3.7
    - nose-exclude==0.5.0
    - psutil==5.6.3
    - pyarrow==0.13.0
    - pymongo==3.8.0
    - querystring-parser==1.2.3
    - seaborn==0.8.1
    - subprocess32==3.5.4
    - tensorboardx==1.7
    - torchvision==0.3.0
    - tqdm==4.32.2
    - websocket-client==0.56.0
prefix: /databricks/python2

GPU 群集上的 Python 2

name: null
channels:
  - Databricks
  - pytorch
  - defaults
dependencies:
  - tensorflow=1.13.1.db1=gpu_py27h8e347d7_0
  - tensorflow-base=1.13.1.db1=gpu_py27he292aa2_0
  - tensorflow-gpu=1.13.1.db1=h0d30ee6_0
  - _libgcc_mutex=0.1=main
  - _py-xgboost-mutex=1.0=gpu_0
  - _tflow_select=2.1.0=gpu
  - absl-py=0.7.1=py27_0
  - asn1crypto=0.24.0=py27_0
  - astor=0.7.1=py27_0
  - backports=1.0=py_2
  - backports.shutil_get_terminal_size=1.0.0=py27_2
  - backports.weakref=1.0.post1=py_1
  - backports_abc=0.5=py_0
  - bcrypt=3.1.6=py27h7b6447c_0
  - blas=1.0=mkl
  - bleach=2.1.3=py27_0
  - boto=2.48.0=py27_1
  - boto3=1.7.62=py27h28b3542_1
  - botocore=1.10.62=py27h28b3542_0
  - ca-certificates=2018.03.07=0
  - certifi=2018.4.16=py27_0
  - cffi=1.11.5=py27he75722e_1
  - chardet=3.0.4=py27_1
  - click=7.0=py27_0
  - cloudpickle=0.8.0=py27_0
  - colorama=0.3.9=py27_0
  - configparser=3.7.3=py27_1
  - cryptography=2.2.2=py27h14c3975_0
  - cudnn=7.6.0=cuda10.0_0
  - cupti=10.0.130=0
  - cycler=0.10.0=py27_0
  - cython=0.28.2=py27h14c3975_0
  - decorator=4.3.0=py27_0
  - docutils=0.14=py27hae222c1_0
  - entrypoints=0.2.3=py27_2
  - enum34=1.1.6=py27_1
  - et_xmlfile=1.0.1=py27h75840f5_0
  - flask=1.0.2=py27_1
  - freetype=2.8=hab7d2ae_1
  - funcsigs=1.0.2=py27_0
  - functools32=3.2.3.2=py27_1
  - future=0.17.1=py27_0
  - futures=3.2.0=py27_0
  - gast=0.2.2=py27_0
  - gitdb2=2.0.5=py27_0
  - gitpython=2.1.11=py27_0
  - gmp=6.1.2=h6c8ec71_1
  - grpcio=1.12.1=py27hdbcaa40_0
  - gunicorn=19.9.0=py27_0
  - h5py=2.8.0=py27h989c5e5_3
  - hdf5=1.10.2=hba1933b_1
  - html5lib=1.0.1=py27_0
  - icu=58.2=h9c2bf20_1
  - idna=2.6=py27h5722d68_1
  - intel-openmp=2018.0.0=8
  - ipaddress=1.0.22=py27_0
  - ipython=5.7.0=py27_0
  - ipython_genutils=0.2.0=py27h89fb69b_0
  - itsdangerous=0.24=py27_1
  - jdcal=1.4=py27_0
  - jinja2=2.10=py27_0
  - jmespath=0.9.4=py_0
  - jpeg=9b=h024ee3a_2
  - jsonschema=2.6.0=py27h7ed5aa4_0
  - jupyter_client=5.2.3=py27_0
  - jupyter_core=4.4.0=py27_0
  - keras=2.2.4=0
  - keras-applications=1.0.8=py_0
  - keras-base=2.2.4=py27_0
  - keras-preprocessing=1.1.0=py_1
  - krb5=1.16.1=hc83ff2d_6
  - libedit=3.1.20170329=h6b74fdf_2
  - libffi=3.2.1=hd88cf55_4
  - libgcc-ng=7.3.0=hdf63c60_0
  - libgfortran-ng=7.2.0=hdf63c60_3
  - libpng=1.6.34=hb9fc6fc_0
  - libpq=10.4=h1ad7b7a_0
  - libprotobuf=3.8.0=hd408876_0
  - libsodium=1.0.16=h1bed415_0
  - libstdcxx-ng=7.3.0=hdf63c60_0
  - libtiff=4.0.9=he85c1e1_2
  - libxgboost=0.90=h688424c_0
  - libxml2=2.9.8=h26e45fe_1
  - libxslt=1.1.32=h1312cb7_0
  - linecache2=1.0.0=py27_0
  - llvmlite=0.23.1=py27hdbcaa40_0
  - lxml=4.2.1=py27h23eabaa_0
  - mako=1.0.10=py_0
  - markdown=3.1.1=py27_0
  - markupsafe=1.0=py27h14c3975_1
  - mistune=0.8.3=py27h14c3975_1
  - mkl=2019.4=243
  - mkl_fft=1.0.12=py27ha843d7b_0
  - mkl_random=1.0.2=py27hd81dba3_0
  - mock=3.0.5=py27_0
  - msgpack-python=0.5.6=py27h6bb024c_1
  - nbconvert=5.3.1=py27_0
  - nbformat=4.4.0=py27hed7f2b2_0
  - ncurses=6.1=he6710b0_1
  - ninja=1.9.0=py27hfd86e86_0
  - numba=0.38.0=py27h637b7d7_0
  - numpy=1.16.2=py27h7e9f1db_0
  - numpy-base=1.16.2=py27hde5b4d6_0
  - olefile=0.45.1=py27_0
  - openpyxl=2.5.3=py27_0
  - openssl=1.0.2o=h14c3975_1
  - pandas=0.23.0=py27h637b7d7_0
  - pandocfilters=1.4.2=py27_1
  - paramiko=2.4.2=py27_0
  - pathlib2=2.3.2=py27_0
  - patsy=0.5.0=py27_0
  - pexpect=4.5.0=py27_0
  - pickleshare=0.7.4=py27h09770e1_0
  - pillow=5.1.0=py27h3deb7b8_0
  - pip=10.0.1=py27_0
  - ply=3.11=py27_0
  - prompt_toolkit=1.0.15=py27_0
  - protobuf=3.8.0=py27he6710b0_0
  - psycopg2=2.7.5=py27hb7f436b_0
  - ptyprocess=0.5.2=py27h4ccb14c_0
  - py-xgboost=0.90=py27h688424c_0
  - py-xgboost-gpu=0.90=py27h28bbb66_0
  - pyasn1=0.4.5=py_0
  - pycparser=2.18=py27_1
  - pygments=2.2.0=py27_0
  - pynacl=1.3.0=py27h7b6447c_0
  - pyopenssl=18.0.0=py27_0
  - pyparsing=2.2.0=py27_1
  - pysocks=1.6.8=py27_0
  - python=2.7.15=h1571d57_0
  - python-dateutil=2.7.3=py27_0
  - python-editor=1.0.4=py_0
  - pytz=2018.4=py27_0
  - pyyaml=5.1=py27h7b6447c_0
  - pyzmq=17.0.0=py27h14c3975_3
  - readline=7.0=h7b6447c_5
  - requests=2.18.4=py27hc5b0589_1
  - s3transfer=0.1.13=py27_0
  - scandir=1.7=py27h14c3975_0
  - scikit-learn=0.20.3=py27hd81dba3_0
  - scipy=1.1.0=py27h7c811a0_2
  - setuptools=39.1.0=py27_0
  - simplegeneric=0.8.1=py27_2
  - simplejson=3.16.0=py27h14c3975_0
  - singledispatch=3.4.0.3=py27h9bcb476_0
  - six=1.11.0=py27_1
  - smmap2=2.0.5=py27_0
  - sqlite=3.23.1=he433501_0
  - sqlparse=0.3.0=py_0
  - statsmodels=0.9.0=py27h035aef0_0
  - tabulate=0.8.3=py27_0
  - tensorboard=1.13.1=py27hf484d3e_0
  - tensorflow-estimator=1.13.0=py_0
  - termcolor=1.1.0=py27_1
  - testpath=0.3.1=py27_0
  - tk=8.6.7=hc745277_3
  - tornado=5.0.2=py27h14c3975_0
  - traceback2=1.4.0=py27_0
  - traitlets=4.3.2=py27hd6ce930_0
  - unittest2=1.1.0=py27_0
  - urllib3=1.22=py27ha55213b_0
  - virtualenv=16.0.0=py27_0
  - wcwidth=0.1.7=py27_0
  - webencodings=0.5.1=py27_1
  - werkzeug=0.14.1=py27_0
  - wheel=0.31.1=py27_0
  - wrapt=1.11.1=py27h7b6447c_0
  - xz=5.2.4=h14c3975_4
  - yaml=0.1.7=had09818_2
  - zeromq=4.2.5=hf484d3e_1
  - zlib=1.2.11=h7b6447c_3
  - pytorch=1.1.0=py2.7_cuda10.0.130_cudnn7.5.1_0
  - torchvision=0.3.0=py27_cu10.0.130_1
  - pip:
    - backports.functools-lru-cache==1.5
    - backports.ssl-match-hostname==3.7.0.1
    - databricks-cli==0.8.7
    - docker==4.0.2
    - fusepy==2.0.4
    - horovod==0.16.4
    - hyperopt==0.1.2.db6
    - kiwisolver==1.1.0
    - matplotlib==2.2.2
    - mleap==0.8.1
    - mlflow==1.0.0
    - msgpack==0.5.6
    - networkx==2.2
    - nose==1.3.7
    - nose-exclude==0.5.0
    - psutil==5.6.3
    - pyarrow==0.13.0
    - pymongo==3.8.0
    - querystring-parser==1.2.3
    - seaborn==0.8.1
    - subprocess32==3.5.4
    - tensorboardx==1.7
    - tqdm==4.32.2
    - websocket-client==0.56.0
prefix: /databricks/python2

包含 Python 模块的 Spark 包

Spark 包 Python 模块 版本
graphframes graphframes 0.7.0-db1-spark2.4
spark-deep-learning sparkdl 1.5.0-db4-spark2.4
tensorframes tensorframes 0.7.0-s_2.11

R 库

R 库与 Databricks Runtime 5.5 中的 R 库完全相同。

Java 和 Scala 库(Scala 2.11 群集)

除了 Databricks Runtime 5.5 中的 Java 库和 Scala 库之外,适用于机器学习的 Databricks Runtime 5.5 LTS 还包含以下 JAR:

组 ID 项目 ID 版本
com.databricks spark-deep-learning 1.5.0-db4-spark2.4
com.typesafe.akka akka-actor_2.11 2.3.11
ml.combust.mleap mleap-databricks-runtime_2.11 0.13.0
ml.dmlc xgboost4j 0.90
ml.dmlc xgboost4j-spark 0.90
org.graphframes graphframes_2.11 0.7.0-db1-spark2.4
org.tensorflow libtensorflow 1.13.1
org.tensorflow libtensorflow_jni 1.13.1
org.tensorflow spark-tensorflow-connector_2.11 1.13.1
org.tensorflow tensorflow 1.13.1
org.tensorframes tensorframes 0.7.0-s_2.11