Databricks Runtime 7.5 for ML(不支持)

Databricks于2020年12月发布了这张图片。

Databricks Runtime 7.5 for Machine Learning为机器学习和数据科学提供了一个现成的环境Databricks Runtime 7.5(不支持).Databricks Runtime ML包含许多流行的机器学习库,包括TensorFlow、PyTorch和XGBoost。它还支持使用Horovod进行分布式深度学习训练。

有关更多信息,包括创建Databricks Runtime ML集群的说明,请参见Databricks运行时机器学习

新特性和主要变化

Databricks Runtime 7.5 ML是建立在Databricks Runtime 7.5之上的。有关Databricks Runtime 7.5(包括Apache Spark MLlib和SparkR)中的新特性的信息,请参见Databricks Runtime 7.5(不支持)发行说明。

的用法

  • 大猩猩包已弃用,并将在即将发布的Databricks Runtime ML主要版本中删除。

Databricks Runtime ML Python环境的主要更改

看到Databricks Runtime 7.5(不支持)参阅Databricks Runtime Python环境的主要更改。有关已安装的Python包及其版本的完整列表,请参见Python库

Python包升级

  • databicks -cli 0.13.0 -> 0.14.0

  • hyperopt 0.2.4。Db2 -> 0.2.5.db1

  • 考拉1.2.0 -> 1.4.0

  • Mlflow 1.11.0 -> 1.12.1

  • 大风暴0.9.6 -> 0.9.7

  • 图4.10.0 -> 4.12.0

  • Pytorch 1.6.0 -> 1.7.0

  • Torchvision 0.7.0 -> 0.8.1

  • Xgboost 1.2.0 -> 1.2.1

系统环境

Databricks Runtime 7.5 ML的系统环境与Databricks Runtime 7.5不同:

以下部分列出了Databricks Runtime 7.5 ML中包含的不同于Databricks Runtime 7.5的库。

Python库

Databricks Runtime 7.5 ML使用Conda进行Python包管理,包括许多流行的ML包。

除了以下章节中Conda环境中指定的包外,Databricks Runtime 7.5 ML还安装了以下包:

  • hyperopt 0.2.5.db1

  • sparkdl 2.1.0-db2

CPU集群上的Python库

的名字databricks-ml渠道-pytorch-违约依赖关系-_libgcc_mutex = 0.1 =主要-absl-py = 0.9.0 = py37_0-= py37_1 1.3.0 asn1crypto =版本-阿斯特= 0.8.0 = py37_0-backcall = 0.1.0 = py37_0-补丁= 1.0 = pyhd3eb1b0_2-bcrypt = 3.2.0 = py37h7b6447c_0-布拉斯特区= 1.0 = mkl-信号灯= 1.4 = py37_0-boto3 = 1.12.0 = py_0-botocore = 1.15.0 = py_0-c-ares = 1.17.1 = h27cfd23_0-ca证书= 2020.10.14 = h06a4308_1#(2021年5月26日维护更新从0更新)-cachetools =以下4.4.1 = py_0-certifi = 2020.11.8 = py37h06a4308_0-cffi = 1.14.0 = py37he30daa8_1#(从2021年5月26日维护更新中的py37h2e261b9_0更新)-chardet = 3.0.4 = py37h06a4308_1003-单击= 7.0 = py37_0-cloudpickle = 1.4.1 = py_0-configparser = 3.7.4 = py37_0-cpuonly = 1.0 = 0-密码= 2.8 = py37h1ba5d50_0-周期计= 0.10.0 = py37_0-cython = 0.29.15 = py37he6710b0_0-4.4.1装饰= = py_0-莳萝= 0.3.1.1 = py37_1-docutils = 0.15.2 = py37_0-entrypoints = 0.3 = py37_0-瓶= 1.1.1 = py_1-freetype的= 2.9.1 = h8a8886c_1-未来= 0.18.2 = py37_1-恐吓= 0.3.3 = py_0-gitdb =你= py_0-gitpython = 3.1.0 = py_0-google-auth = 1.11.2 = py_0-google-auth-oauthlib = 0.4.1 = py_2-google-pasta = 0.2.0 = py_0-grpcio = 1.27.2 = py37hf8bcb03_0-gunicorn = 20.0.4 = py37_0-h5py = 2.10.0 = py37h7918eee_0-hdf5 = 1.10.4 = hb1b8bf9_0-icu = 58.2 = he6710b0_3-idna = 2.8 = py37_0-intel-openmp = 2020.0 = 166-ipykernel = 5.1.4 = py37h39e3cac_0-ipython = 7.12.0 = py37h5ca1d4c_0-ipython_genutils = 0.2.0 = pyhd3eb1b0_1-isodate = 0.6.0 = py_1-itsdangerous = 1.1.0 = py37_0-绝地= 0.17.2 = py37_0-jinja2 = 2.11.1 = py_0-jmespath = 0.10.0 = py_0-joblib = 0.14.1 = py_0-jpeg = 9 b = h024ee3a_2-jupyter_client = 5.3.4 = py37_0-jupyter_core = 4.6.1 = py37_0-kiwisolver = 1.1.0 = py37he6710b0_0-krb5 = 1.17.1 = h173b8e3_0#(2021年5月26日维护更新从1.16.4更新)-ld_impl_linux - 64 = 2.33.1 = h53a641e_7-libedit = 3.1.20181209 = hc058e9b_0-libffi = 3.3 = he6710b0_2#(2021年5月26日维护更新从3.2.1更新)-libgcc-ng = 9.1.0 = hdf63c60_0-libgfortran-ng = 7.3.0 = hdf63c60_0-libpng = 1.6.37 = hbc83047_0-libpq = 12.2 = h20c2e04_0#(在2021年5月26日的维护更新中从11.2更新)-libprotobuf = 3.11.4 = hd408876_0-libsodium = 1.0.16 = h1bed415_0-libstdcxx-ng = 9.1.0 = hdf63c60_0-libtiff = 4.1.0 = h2733197_0-libuv = 1.40.0 = h7b6447c_0-lightgbm = tripwire = py37he6710b0_0-lz4-c = 1.8.1.2 = h14c3975_0-尖吻鲭鲨= 1.1.2 = py_0-减价= 3.1.1 = py37_0-markupsafe = 1.1.1 = py37h14c3975_1-matplotlib-base = 3.1.3 = py37hef1b27d_0-mkl = 2020.0 = 166-mkl-service = tripwire = py37he904b0f_0-mkl_fft = 1.0.15 = py37ha843d7b_0-mkl_random = 1.1.0 = py37hd6b4f25_0-ncurses = 6.2 = he6710b0_1-networkx = 2.4 = py_1-忍者= 1.10.2 = py37hff7bd54_0-nltk = 3.4.5 = py37_0-numpy = 1.18.1 = py37h4f9e942_0-numpy-base = 1.18.1 = py37hde5b4d6_1-oauthlib = 3.1.0 = py_0-olefile = 0.46 = py37_0-openssl = 1.1.1k = h27cfd23_0#(从2021年5月26日维护更新的1.1.1h更新)-包装= 20.1 = py_0-熊猫= 1.0.1 = py37h0573a6f_0-paramiko = 2.7.1 = py_0-parso = 0.7.0 = py_0-容易受骗的人= 0.5.1 = py37_0-pexpect = 4.8.0 = pyhd3eb1b0_3-pickleshare = 0.7.5 = py37_1001-枕头= 7.0.0 = py37hb39fc2d_0-皮普= 20.0.2 = py37_3-情节= 4.12.0 = pyhd3eb1b0_0-prompt_toolkit = 3.0.3 = py_0-protobuf = 3.11.4 = py37he6710b0_0-psutil = 5.6.7 = py37h7b6447c_0-psycopg2 = 2.8.6 = py37h3c74f83_1#(2021年5月26日维护更新从2.8.4更新)-ptyprocess = 0.6.0 = pyhd3eb1b0_2-pyasn1 = 0.4.8 = py_0-pyasn1-modules = 0.2.8 = py_0-pycparser = 2.19 = py37_0-pygments = 2.5.2 = py_0-pyjwt = 1.7.1上= py37_0-= py37h7b6447c_0 1.3.0 pynacl =版本-pyodbc = 4.0.30 = py37he6710b0_0-pyopenssl = 19.1.0 = pyhd3eb1b0_1-pyparsing = 2.4.6 = py_0-pysocks = 1.7.1上= py37_1-python = 3.7.10 = hdb3f193_0#(2021年5月26日维护更新从3.7.6更新)-python-dateutil = 2.8.1发布= py_0-python编辑器的1.0.4 = = py_0-pytorch = 1.7.0 = py3.7_cpu_0-pytz = 2019.3 = py_0-pyzmq = 18.1.1 = py37he6710b0_0-readline = 8.1 = h27cfd23_0#(更新自7.0在2021年5月26日维护更新)-= 2.22.0 = py37_1请求-= py_0 1.3.0 requests-oauthlib =版本-重试= 1.3.3 = py37_2-rsa = 4.0 = py_0-s3transfer = 0.3.3 = py37_1-scikit-learn = 0.22.1 = py37hd81dba3_0-scipy = 1.4.1 = py37h0b6359f_0-setuptools = 45.2.0 = py37_0-simplejson = 3.17.0 = py37h7b6447c_0-6 = 1.14.0 = py37h06a4308_0-smmap = 3.0.4 = py_0-sqlite = 3.35.4 = hdfb4753_0#(2021年5月26日维护更新从3.31.1更新)-sqlparse = 0.4.1 = py_0-statsmodels = 0.11.0 = py37h7b6447c_0它-汇总= 0.8.3 = py37_0-tk = 8.6.10 = hbc83047_0#(2021年5月26日维护更新从8.6.8更新)-torchvision = 0.8.1 = py37_cpu-龙卷风= 6.0.3 = py37h7b6447c_3-tqdm = 4.42.1 = py_0-traitlets = 4.3.3 = py37_0-typing_extensions = 3.7.4.3 = py_0-unixodbc = 2.3.7 = h14c3975_0-urllib3 = 1.25.8 = py37_0-wcwidth = 0.1.8 = py_0-websocket-client = 0.56.0 = py37_0-werkzeug = 1.0.0 = py_0-轮= 0.34.2 = py37_0-打包= 1.11.2 = py37h7b6447c_0-xz = 5.2.5 = h7b6447c_0#(2021年5月26日维护更新从5.2.4更新)-zeromq = 4.3.1 = he6710b0_3-zlib = 1.2.11 = h7b6447c_3-zstd = 1.3.7 = h0b5b093_0-皮普-astunparse = = 1.6.3-azure-core = = 1.9.0-azure-storage-blob = = 12.6.0-databricks-cli = = 0.14.0-diskcache = = 5.1.0-码头工人= = 4.4.0-大猩猩= = 0.3.0-horovod = = 0.20.3-joblibspark = = 0.3.0-keras-preprocessing = = 1.1.2-考拉= = 1.4.0-mleap = = 0.16.1-mlflow = = 1.12.1-msr = = 0.6.19-opt-einsum = = 3.3.0-petastorm = = 0.9.7-pyarrow = = 1.0.1-pyyaml = = 5.3.1-querystring-parser = = 4-seaborn = = 0.10.0-spark-tensorflow-distributor = = 0.1.0-tensorboard = = tripwire-tensorboard-plugin-wit = = 1.7.0-tensorflow-cpu = = 2.3.1-tensorflow-estimator = = tripwire-termcolor = = 1.1.0-xgboost = = 1.2.1 "前缀/砖/ conda / env / databricks-ml

GPU集群上的Python库

的名字databricks-ml-gpu渠道-pytorch-违约依赖关系-_libgcc_mutex = 0.1 =主要-absl-py = 0.9.0 = py37_0-= py37_1 1.3.0 asn1crypto =版本-阿斯特= 0.8.0 = py37_0-backcall = 0.1.0 = py37_0-补丁= 1.0 = pyhd3eb1b0_2-bcrypt = 3.2.0 = py37h7b6447c_0-布拉斯特区= 1.0 = mkl-信号灯= 1.4 = py37_0-boto3 = 1.12.0 = py_0-botocore = 1.15.0 = py_0-c-ares = 1.17.1 = h27cfd23_0-ca证书= 2020.10.14 = h06a4308_1#(2021年5月26日维护更新从0更新)-cachetools =以下4.4.1 = py_0-certifi = 2020.11.8 = py37h06a4308_0-cffi = 1.14.0 = py37he30daa8_1#(从2021年5月26日维护更新中的py37h2e261b9_0更新)-chardet = 3.0.4 = py37h06a4308_1003-单击= 7.0 = py37_0-cloudpickle = 1.4.1 = py_0-configparser = 3.7.4 = py37_0-密码= 2.8 = py37h1ba5d50_0-cudatoolkit = 10.1.243 = h6bb024c_0-周期计= 0.10.0 = py37_0-cython = 0.29.15 = py37he6710b0_0-4.4.1装饰= = py_0-莳萝= 0.3.1.1 = py37_1-docutils = 0.15.2 = py37_0-entrypoints = 0.3 = py37_0-瓶= 1.1.1 = py_1-freetype的= 2.9.1 = h8a8886c_1-未来= 0.18.2 = py37_1-恐吓= 0.3.3 = py_0-gitdb =你= py_0-gitpython = 3.1.0 = py_0-google-auth = 1.11.2 = py_0-google-auth-oauthlib = 0.4.1 = py_2-google-pasta = 0.2.0 = py_0-grpcio = 1.27.2 = py37hf8bcb03_0-gunicorn = 20.0.4 = py37_0-h5py = 2.10.0 = py37h7918eee_0-hdf5 = 1.10.4 = hb1b8bf9_0-icu = 58.2 = he6710b0_3-idna = 2.8 = py37_0-intel-openmp = 2020.0 = 166-ipykernel = 5.1.4 = py37h39e3cac_0-ipython = 7.12.0 = py37h5ca1d4c_0-ipython_genutils = 0.2.0 = pyhd3eb1b0_1-isodate = 0.6.0 = py_1-itsdangerous = 1.1.0 = py37_0-绝地= 0.17.2 = py37_0-jinja2 = 2.11.1 = py_0-jmespath = 0.10.0 = py_0-joblib = 0.14.1 = py_0-jpeg = 9 b = h024ee3a_2-jupyter_client = 5.3.4 = py37_0-jupyter_core = 4.6.1 = py37_0-kiwisolver = 1.1.0 = py37he6710b0_0-krb5 = 1.17.1 = h173b8e3_0#(2021年5月26日维护更新从1.16.4更新)-ld_impl_linux - 64 = 2.33.1 = h53a641e_7-libedit = 3.1.20181209 = hc058e9b_0-libffi = 3.3 = he6710b0_2#(2021年5月26日维护更新从3.2.1更新)-libgcc-ng = 9.1.0 = hdf63c60_0-libgfortran-ng = 7.3.0 = hdf63c60_0-libpng = 1.6.37 = hbc83047_0-libpq = 12.2 = h20c2e04_0#(在2021年5月26日的维护更新中从11.2更新)-libprotobuf = 3.11.4 = hd408876_0-libsodium = 1.0.16 = h1bed415_0-libstdcxx-ng = 9.1.0 = hdf63c60_0-libtiff = 4.1.0 = h2733197_0-libuv = 1.40.0 = h7b6447c_0-lightgbm = tripwire = py37he6710b0_0-lz4-c = 1.8.1.2 = h14c3975_0-尖吻鲭鲨= 1.1.2 = py_0-减价= 3.1.1 = py37_0-markupsafe = 1.1.1 = py37h14c3975_1-matplotlib-base = 3.1.3 = py37hef1b27d_0-mkl = 2020.0 = 166-mkl-service = tripwire = py37he904b0f_0-mkl_fft = 1.0.15 = py37ha843d7b_0-mkl_random = 1.1.0 = py37hd6b4f25_0-ncurses = 6.2 = he6710b0_1-networkx = 2.4 = py_1-忍者= 1.10.2 = py37hff7bd54_0-nltk = 3.4.5 = py37_0-numpy = 1.18.1 = py37h4f9e942_0-numpy-base = 1.18.1 = py37hde5b4d6_1-oauthlib = 3.1.0 = py_0-olefile = 0.46 = py37_0-openssl = 1.1.1k = h27cfd23_0#(从2021年5月26日维护更新的1.1.1h更新)-包装= 20.1 = py_0-熊猫= 1.0.1 = py37h0573a6f_0-paramiko = 2.7.1 = py_0-parso = 0.7.0 = py_0-容易受骗的人= 0.5.1 = py37_0-pexpect = 4.8.0 = pyhd3eb1b0_3-pickleshare = 0.7.5 = py37_1001-枕头= 7.0.0 = py37hb39fc2d_0-皮普= 20.0.2 = py37_3-情节= 4.12.0 = pyhd3eb1b0_0-prompt_toolkit = 3.0.3 = py_0-protobuf = 3.11.4 = py37he6710b0_0-psutil = 5.6.7 = py37h7b6447c_0-psycopg2 = 2.8.6 = py37h3c74f83_1#(2021年5月26日维护更新从2.8.4更新)-ptyprocess = 0.6.0 = pyhd3eb1b0_2-pyasn1 = 0.4.8 = py_0-pyasn1-modules = 0.2.8 = py_0-pycparser = 2.19 = py37_0-pygments = 2.5.2 = py_0-pyjwt = 1.7.1上= py37_0-= py37h7b6447c_0 1.3.0 pynacl =版本-pyodbc = 4.0.30 = py37he6710b0_0-pyopenssl = 19.1.0 = pyhd3eb1b0_1-pyparsing = 2.4.6 = py_0-pysocks = 1.7.1上= py37_1-python = 3.7.10 = hdb3f193_0#(2021年5月26日维护更新从3.7.6更新)-python-dateutil = 2.8.1发布= py_0-python编辑器的1.0.4 = = py_0-pytorch = 1.7.0 = py3.7_cuda10.1.243_cudnn7.6.3_0-pytz = 2019.3 = py_0-pyzmq = 18.1.1 = py37he6710b0_0-readline = 8.1 = h27cfd23_0#(更新自7.0在2021年5月26日维护更新)-= 2.22.0 = py37_1请求-= py_0 1.3.0 requests-oauthlib =版本-重试= 1.3.3 = py37_2-rsa = 4.0 = py_0-s3transfer = 0.3.3 = py37_1-scikit-learn = 0.22.1 = py37hd81dba3_0-scipy = 1.4.1 = py37h0b6359f_0-setuptools = 45.2.0 = py37_0-simplejson = 3.17.0 = py37h7b6447c_0-6 = 1.14.0 = py37h06a4308_0-smmap = 3.0.4 = py_0-sqlite = 3.35.4 = hdfb4753_0#(2021年5月26日维护更新从3.31.1更新)-sqlparse = 0.4.1 = py_0-statsmodels = 0.11.0 = py37h7b6447c_0它-汇总= 0.8.3 = py37_0-tk = 8.6.10 = hbc83047_0#(2021年5月26日维护更新从8.6.8更新)-torchvision = 0.8.1 = py37_cu101-龙卷风= 6.0.3 = py37h7b6447c_3-tqdm = 4.42.1 = py_0-traitlets = 4.3.3 = py37_0-typing_extensions = 3.7.4.3 = py_0-unixodbc = 2.3.7 = h14c3975_0-urllib3 = 1.25.8 = py37_0-wcwidth = 0.1.8 = py_0-websocket-client = 0.56.0 = py37_0-werkzeug = 1.0.0 = py_0-轮= 0.34.2 = py37_0-打包= 1.11.2 = py37h7b6447c_0-xz = 5.2.5 = h7b6447c_0#(2021年5月26日维护更新从5.2.4更新)-zeromq = 4.3.1 = he6710b0_3-zlib = 1.2.11 = h7b6447c_3-zstd = 1.3.7 = h0b5b093_0-皮普-astunparse = = 1.6.3-azure-core = = 1.9.0-azure-storage-blob = = 12.6.0-databricks-cli = = 0.14.0-diskcache = = 5.1.0-码头工人= = 4.4.0-大猩猩= = 0.3.0-horovod = = 0.20.3-joblibspark = = 0.3.0-keras-preprocessing = = 1.1.2-考拉= = 1.4.0-mleap = = 0.16.1-mlflow = = 1.12.1-msr = = 0.6.19-opt-einsum = = 3.3.0-petastorm = = 0.9.7-pyarrow = = 1.0.1-pyyaml = = 5.3.1-querystring-parser = = 4-seaborn = = 0.10.0-spark-tensorflow-distributor = = 0.1.0-tensorboard = = tripwire-tensorboard-plugin-wit = = 1.7.0-tensorflow = = 2.3.1-tensorflow-estimator = = tripwire-termcolor = = 1.1.0-xgboost = = 1.2.1 "前缀/砖/ conda / env / databricks-ml-gpu

包含Python模块的Spark包

火花包

Python模块

版本

graphframes

graphframes

0.8.1-db1-spark3.0

R库

R库与R库在Databricks运行时7.5。

Java和Scala库(Scala 2.12集群)

除了Java和Scala库在Databricks Runtime 7.5, Databricks Runtime 7.5 ML包含以下jar:

CPU集群

组ID

工件ID

版本

com.typesafe.akka

akka-actor_2.12

2.5.23

ml.combust.mleap

mleap-databricks-runtime_2.12

0.17.3-4882dc3

ml.dmlc

xgboost4j-spark_2.12

1.2.0

ml.dmlc

xgboost4j_2.12

1.2.0

org.mlflow

mlflow-client

1.12.1

org.scala-lang.modules

scala-java8-compat_2.12

0.8.0

org.tensorflow

spark-tensorflow-connector_2.12

1.15.0

GPU集群

组ID

工件ID

版本

com.typesafe.akka

akka-actor_2.12

2.5.23

ml.combust.mleap

mleap-databricks-runtime_2.12

0.17.3-4882dc3

ml.dmlc

xgboost4j-spark-gpu_2.12

1.2.0

ml.dmlc

xgboost4j-gpu_2.12

1.2.0

org.mlflow

mlflow-client

1.12.1

org.scala-lang.modules

scala-java8-compat_2.12

0.8.0

org.tensorflow

spark-tensorflow-connector_2.12

1.15.0