Partager via

Databricks Runtime 5.5 LTS pour ML (EoS)

  • Article

Remarque

La prise en charge de cette version databricks Runtime a pris fin. Pour connaître la date de fin de support, consultez l’historique de fin de support. Pour toutes les versions prises en charge de Databricks Runtime, consultez Notes de publication sur les versions et la compatibilité de Databricks Runtime.

Databricks a publié cette version en juillet 2019. Le support a pris fin le 27 juillet 2021. Databricks Runtime 5.5 ML Support étendu (EoS) prolonge le support de la version 5.5 ML jusqu’à décembre 2021. Il utilise Ubuntu 18.04.5 LTS au lieu de la distribution Ubuntu 16.04.6 LTS déconseillée utilisée dans la version originale Databricks Runtime 5.5 ML LTS. Le support d’Ubuntu 16.04.6 LTS a cessé le 1er avril 2021.

Databricks Runtime 5.5 LTS pour Machine Learning fournit un environnement prêt à l’emploi pour Machine Learning et la science des données basé sur Databricks Runtime 5.5 LTS (EoS). Databricks Runtime ML contient de nombreuses bibliothèques populaires de Machine Learning, notamment TensorFlow, PyTorch, Keras et XGBoost. Il prend également en charge la formation de Deep Learning distribué avec Horovod.

Pour plus d’informations, notamment les instructions relatives à la création d’un groupement Databricks Runtime ML, consultez IA et apprentissage automatique sur Databricks.

Nouvelles fonctionnalités

Databricks Runtime 5.5 LTS pour le Machine Learning repose sur Databricks Runtime 5.5 LTS. Pour plus d’informations sur les nouveautés de Databricks Runtime 5.5 LTS, consultez les notes de publication de Databricks Runtime 5.5 LTS (EoS).

En plus des mises à jour de bibliothèque, Databricks Runtime 5.5 LTS pour le Machine Learning introduit les nouvelles fonctionnalités suivantes :

  • Ajout du package Python MLflow 1.0

Améliorations

  • Mise à niveau des bibliothèques de Machine Learning

    • TensorFlow mise à niveau de 1.12.0 vers 1.13.1
    • PyTorch mise à niveau de 0.4.1 vers 1.1.0
    • scikit-learn mise à niveau de 0.19.1 vers 0.20.3

  • Opération à un seul nœud pour HorovodRunner

    Activation de HorovodRunner pour l’exécution uniquement sur le nœud de pilote. Auparavant, pour utiliser HorovodRunner, vous deviez exécuter un pilote et au moins un nœud Worker. Avec cette modification, vous pouvez désormais distribuer l’apprentissage au sein d’un nœud unique (autrement dit, un nœud à plusieurs GPU) et ainsi utiliser les ressources de calcul plus efficacement.

Dépréciation

Dans la bibliothèque Hyperopt, nous avons déconseillé les propriétés suivantes de hyperopt.SparkTrials :

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

et remplacé les propriétés par les fonctions suivantes :

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

Environnement du système

L’environnement système de Databricks Runtime 5.5 LTS pour le Machine Learning diffère de Databricks Runtime 5.5 comme suit :

Bibliothèques

Les sections suivantes répertorient les bibliothèques incluses dans Databricks Runtime 5.5 LTS pour le Machine Learning qui diffèrent de celles incluses dans Databricks Runtime 5.5.

Bibliothèques de niveau supérieur

Databricks Runtime 5.5 LTS pour le Machine Learning comprend les bibliothèques de niveau supérieur suivantes :

Bibliothèques Python

Databricks Runtime 5.5 LTS pour le Machine Learning utilise Conda pour la gestion de package Python. Par conséquent, il existe des différences majeures dans les bibliothèques Python installées par rapport à Databricks Runtime. Les sections suivantes décrivent les environnements Conda pour clusters Databricks Runtime 5.5 LTS pour le Machine Learning utilisant Python 2 ou 3, ainsi que pour machines virtuelles avec UC et GPU.

Python 3 sur les clusters UC

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

Python 3 sur les clusters GPU

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

Python 2 sur les clusters UC

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

Python 2 sur les clusters GPU

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

Packages Spark contenant des modules Python

Package Spark Module Python Version
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

Bibliothèques R

Les bibliothèques R sont identiques aux bibliothèques R dans Databricks Runtime 5.5.

Bibliothèques Java et Scala (cluster Scala 2.11)

En plus des bibliothèques Java et Scala dans Databricks Runtime 5.5, Databricks Runtime 5.5 LTS pour le Machine Learning contient les fichiers JAR suivants :

ID de groupe ID d’artefact Version
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