-
Notifications
You must be signed in to change notification settings - Fork 86
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Merge branch 'master' into master-dl
- Loading branch information
Showing
50 changed files
with
148 additions
and
77 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Binary file added
BIN
+840 Bytes
...om/upload/abalone.csv/.part-00001-ef41bb04-d8a6-4b5a-911d-59300b26dfe1.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+104 KB
...ung.com/upload/abalone.csv/part-00001-ef41bb04-d8a6-4b5a-911d-59300b26dfe1.snappy.parquet
Binary file not shown.
Binary file added
BIN
+300 Bytes
...com/upload/income.csv/.part-00001-f1b4c004-b653-4ef7-beb6-2b18d482d48f.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+36.2 KB
...sung.com/upload/income.csv/part-00001-f1b4c004-b653-4ef7-beb6-2b18d482d48f.snappy.parquet
Binary file not shown.
Binary file added
BIN
+128 Bytes
...d/sample_ALSTrain.csv/.part-00001-41f1abfa-afd1-4aff-a24f-b4ff0306fff3.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+14.8 KB
...upload/sample_ALSTrain.csv/part-00001-41f1abfa-afd1-4aff-a24f-b4ff0306fff3.snappy.parquet
Binary file not shown.
Binary file added
BIN
+16 Bytes
...e_AssociationRule.csv/.part-00001-2f3104d2-c1bc-4de5-ac2b-30428195e3fd.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+545 Bytes
...sample_AssociationRule.csv/part-00001-2f3104d2-c1bc-4de5-ac2b-30428195e3fd.snappy.parquet
Binary file not shown.
Binary file added
BIN
+16 Bytes
.../sample_Binarizer.csv/.part-00001-4893a240-ddd2-417a-923d-a2f2ca520613.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+680 Bytes
...pload/sample_Binarizer.csv/part-00001-4893a240-ddd2-417a-923d-a2f2ca520613.snappy.parquet
Binary file not shown.
Binary file added
BIN
+48 Bytes
...upload/sample_GLM.csv/.part-00001-fc9fa294-5b4c-40e5-a1f3-0a7f08641818.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+4.53 KB
....com/upload/sample_GLM.csv/part-00001-fc9fa294-5b4c-40e5-a1f3-0a7f08641818.snappy.parquet
Binary file not shown.
Binary file added
BIN
+12 Bytes
...ensityEstimation2.csv/.part-00001-4e166ce8-312e-44a6-bae5-b9536146c6d1.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+346 Bytes
...rnelDensityEstimation2.csv/part-00001-4e166ce8-312e-44a6-bae5-b9536146c6d1.snappy.parquet
Binary file not shown.
Binary file added
BIN
+16 Bytes
...richletAllocation.csv/.part-00001-b4a777a3-f7ba-4ca7-8557-e2b4e7ac643d.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+549 Bytes
...entDirichletAllocation.csv/part-00001-b4a777a3-f7ba-4ca7-8557-e2b4e7ac643d.snappy.parquet
Binary file not shown.
Binary file added
BIN
+16 Bytes
...load/sample_Ngram.csv/.part-00001-4e58c336-1aca-49f1-98b0-66b865693af0.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+797 Bytes
...om/upload/sample_Ngram.csv/part-00001-4e58c336-1aca-49f1-98b0-66b865693af0.snappy.parquet
Binary file not shown.
Binary file added
BIN
+16 Bytes
...ample_StringSplit.csv/.part-00001-50d40c7b-69b6-41c5-92d7-4449c5f2cddd.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+682 Bytes
...oad/sample_StringSplit.csv/part-00001-50d40c7b-69b6-41c5-92d7-4449c5f2cddd.snappy.parquet
Binary file not shown.
Binary file added
BIN
+5.8 KB
...load/sample_adult.csv/.part-00001-43f35916-5eb0-4856-9ec5-430e4fcd2533.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+741 KB
...om/upload/sample_adult.csv/part-00001-43f35916-5eb0-4856-9ec5-430e4fcd2533.snappy.parquet
Binary file not shown.
Binary file added
BIN
+36 Bytes
...ple_airPassengers.csv/.part-00001-60de8c4a-aa05-446a-8db6-c6bdec01d0bf.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+3.38 KB
...d/sample_airPassengers.csv/part-00001-60de8c4a-aa05-446a-8db6-c6bdec01d0bf.snappy.parquet
Binary file not shown.
Binary file added
BIN
+20 Bytes
...pload/sample_data.csv/.part-00001-b5493373-d07f-418a-bd75-43962f77afcc.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+1.38 KB
...com/upload/sample_data.csv/part-00001-b5493373-d07f-418a-bd75-43962f77afcc.snappy.parquet
Binary file not shown.
Binary file added
BIN
+160 Bytes
...ata_tokenizer_eng.csv/.part-00001-e420b6a8-0f2b-46c6-94f4-f9aa90c7a6d1.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+18.5 KB
...ple_data_tokenizer_eng.csv/part-00001-e420b6a8-0f2b-46c6-94f4-f9aa90c7a6d1.snappy.parquet
Binary file not shown.
Binary file added
BIN
+20 Bytes
...d/sample_datetime.csv/.part-00001-1869d30c-3ba7-442a-96cc-96e39a1f27cc.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+1.29 KB
...upload/sample_datetime.csv/part-00001-1869d30c-3ba7-442a-96cc-96e39a1f27cc.snappy.parquet
Binary file not shown.
Binary file added
BIN
+528 Bytes
...pload/sample_nsmc.csv/.part-00001-9b50e2fe-452c-4700-9f0a-0ea3b2e28de5.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+64.8 KB
...com/upload/sample_nsmc.csv/part-00001-9b50e2fe-452c-4700-9f0a-0ea3b2e28de5.snappy.parquet
Binary file not shown.
Binary file added
BIN
+7.6 KB
...mple_termDocument.csv/.part-00001-d9599308-04dc-448f-b6fd-62016bf9f209.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+972 KB
...ad/sample_termDocument.csv/part-00001-d9599308-04dc-448f-b6fd-62016bf9f209.snappy.parquet
Binary file not shown.
Binary file added
BIN
+24 Bytes
...sample_timeseries.csv/.part-00001-5a62a26c-573d-4414-916e-e389af3b9bdb.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+1.71 KB
...load/sample_timeseries.csv/part-00001-5a62a26c-573d-4414-916e-e389af3b9bdb.snappy.parquet
Binary file not shown.
Binary file added
BIN
+20 Bytes
...om/upload/script1.csv/.part-00001-8c37e7cf-7b5f-4bfe-98d2-6bb3d2c04b17.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+1.18 KB
...ung.com/upload/script1.csv/part-00001-8c37e7cf-7b5f-4bfe-98d2-6bb3d2c04b17.snappy.parquet
Binary file not shown.
Binary file added
BIN
+4.59 KB
...d/stateoftheunion.csv/.part-00001-811a5401-d02b-4599-9d3a-e89b90372bc8.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+586 KB
...upload/stateoftheunion.csv/part-00001-811a5401-d02b-4599-9d3a-e89b90372bc8.snappy.parquet
Binary file not shown.
Binary file added
BIN
+32 Bytes
...m/upload/text_eng.csv/.part-00001-948b463e-c4a8-4222-aa9e-f1f899fce781.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+2.87 KB
...ng.com/upload/text_eng.csv/part-00001-948b463e-c4a8-4222-aa9e-f1f899fce781.snappy.parquet
Binary file not shown.
Binary file added
BIN
+12 Bytes
.../upload/user_dict.csv/.part-00001-420db257-1ab6-4cb6-baf9-aa6b7e25ed27.snappy.parquet.crc
Binary file not shown.
Binary file added
BIN
+429 Bytes
...g.com/upload/user_dict.csv/part-00001-420db257-1ab6-4cb6-baf9-aa6b7e25ed27.snappy.parquet
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,43 +1,62 @@ | ||
FROM centos:7 | ||
# Stage 1: builder | ||
FROM centos:centos7 as builder | ||
|
||
WORKDIR /opt | ||
|
||
# set basic environment | ||
RUN yum install -y git java-1.8.0-openjdk-devel bzip2 \ | ||
&& yum groupinstall -y "Development Tools" \ | ||
&& curl -s http://apache.mirror.cdnetworks.com/maven/maven-3/3.6.0/binaries/apache-maven-3.6.0-bin.tar.gz | tar xzv \ | ||
&& curl -s https://nodejs.org/download/release/v8.11.2/node-v8.11.2-linux-x64.tar.gz | tar xzv \ | ||
&& curl -LO https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh \ | ||
&& sh Miniconda3-latest-Linux-x86_64.sh -b -p /opt/miniconda3 | ||
|
||
ENV PYTHON_HOME=/opt/miniconda3 \ | ||
JAVA_HOME=/usr/lib/jvm/java \ | ||
NODEJS_HOME=/opt/node-v8.11.2-linux-x64 \ | ||
M2_HOME=/opt/apache-maven-3.6.0 | ||
|
||
RUN yum install -y git java-1.8.0-openjdk-devel bzip2 | ||
RUN yum groupinstall -y "Development Tools" | ||
RUN curl -s http://apache.mirror.cdnetworks.com/maven/maven-3/3.6.3/binaries/apache-maven-3.6.3-bin.tar.gz | tar xzv | ||
RUN curl -s https://nodejs.org/download/release/v8.11.2/node-v8.11.2-linux-x64.tar.gz | tar xzv | ||
RUN curl -LO https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh | ||
RUN sh Miniconda3-latest-Linux-x86_64.sh -b -p /opt/miniconda3 | ||
|
||
ENV PYTHON_HOME=/opt/miniconda3 | ||
ENV JAVA_HOME=/usr/lib/jvm/java | ||
ENV NODEJS_HOME=/opt/node-v8.11.2-linux-x64 | ||
ENV M2_HOME=/opt/apache-maven-3.6.3 | ||
ENV PATH=$PYTHON_HOME/bin:$JAVA_HOME/bin:$NODEJS_HOME/bin:${M2_HOME}/bin:$PATH | ||
|
||
|
||
# clone and package | ||
RUN mkdir /git \ | ||
&& cd /git \ | ||
&& git clone https://github.com/brightics/studio.git \ | ||
&& cd /git/studio \ | ||
&& mvn clean package -DskipTests | ||
|
||
RUN mkdir /git | ||
RUN cd /git && git clone https://github.com/brightics/studio.git | ||
RUN cd /git/studio && mvn clean package -DskipTests | ||
|
||
# setup | ||
WORKDIR /brightics-studio | ||
RUN mv /git/studio/build/target/dist/brightics-studio / \ | ||
&& cd /brightics-studio \ | ||
&& sed -i "s/\%\*\ //g" setup.sh \ | ||
&& ./setup.sh | ||
|
||
RUN mv /git/studio/build/target/dist/brightics-studio / | ||
RUN sed -i "s/\"127.0.0.1\",/\"0.0.0.0\",/g" /brightics-studio/visual-analytics/conf.json | ||
RUN sed -i "s/\%\*\ //g" /brightics-studio/setup.sh | ||
RUN sed -i '24d' /brightics-studio/start-brightics.sh | ||
RUN ./setup.sh | ||
RUN rm -rf /brightics-studio/lib/etc /brightics-studio/lib/graphviz /brightics-studio/lib/hadoop /brightics-studio/lib/shortcut /brightics-studio/lib/node/node_modules/npm/changelogs /brightics-studio/lib/node/node_modules/npm/doc /brightics-studio/lib/node/node_modules/npm/html /brightics-studio/lib/node/node_modules/npm/man /brightics-studio/lib/node/node_modules/npm/scripts /brightics-studio/lib/node/node_modules/npm/*.md /brightics-studio/lib/node/node_modules/npm/AUTHORS /brightics-studio/lib/node/node_modules/npm/TODO.org /brightics-studio/lib/node/node_modules/npm/.github /opt/zulu8.46.0.19-ca-jre8.0.252-linux_x64/man | ||
|
||
|
||
# Stage 2 | ||
FROM centos:centos7 | ||
|
||
COPY --from=builder /brightics-studio /brightics-studio | ||
COPY --from=builder /opt/miniconda3 /opt/miniconda3 | ||
|
||
WORKDIR /opt | ||
|
||
RUN yum install -y graphviz && \ | ||
yum clean all && \ | ||
rm -rf /var/cache/yum && \ | ||
curl -s https://cdn.azul.com/zulu/bin/zulu8.46.0.19-ca-jre8.0.252-linux_x64.tar.gz | tar xzv | ||
|
||
ENV PYTHON_HOME=/opt/miniconda3 \ | ||
JAVA_HOME=/opt/zulu8.46.0.19-ca-jre8.0.252-linux_x64 \ | ||
NODEJS_HOME=/brightics-studio/lib/node | ||
|
||
ENV PATH=$PYTHON_HOME/bin:$JAVA_HOME/bin:$NODEJS_HOME:$PATH | ||
|
||
WORKDIR /brightics-studio | ||
|
||
# expose the port for visual-analytics | ||
EXPOSE 3000 | ||
|
||
|
||
ADD entrypoint.sh /etc/entrypoint.sh | ||
RUN chmod +x /etc/entrypoint.sh | ||
ENTRYPOINT ["/etc/entrypoint.sh"] |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,44 +1,81 @@ | ||
arch | ||
asn1crypto | ||
boto3 | ||
botocore | ||
cached-property | ||
cffi | ||
cryptography | ||
cx-Oracle | ||
cycler | ||
Cython | ||
docutils | ||
graphviz | ||
idna | ||
jmespath | ||
kiwisolver | ||
lightgbm | ||
matplotlib | ||
numpy | ||
pandas | ||
pandasql | ||
patsy | ||
pg8000 | ||
py4j | ||
pyarrow | ||
pycparser | ||
pydotplus | ||
PyMySQL | ||
pyparsing | ||
python-dateutil | ||
pytz | ||
redis | ||
s3transfer | ||
scikit-learn | ||
scipy | ||
seaborn | ||
six | ||
sklearn | ||
SQLAlchemy | ||
tqdm | ||
urllib3 | ||
xgboost | ||
statsmodels | ||
pandas-profiling | ||
numexpr==2.6.4 | ||
arch==4.7.0 | ||
asn1crypto==0.24.0 | ||
boto==2.49.0 | ||
boto3==1.9.67 | ||
botocore==1.12.67 | ||
cached-property==1.5.1 | ||
certifi==2018.11.29 | ||
cffi==1.11.5 | ||
chardet==3.0.4 | ||
cryptography==2.4.2 | ||
cx-Oracle==7.0.0 | ||
cycler==0.10.0 | ||
Cython==0.29.2 | ||
decorator==4.3.0 | ||
docker==3.7.0 | ||
docker-compose==1.23.2 | ||
docker-pycreds==0.4.0 | ||
dockerpty==0.4.1 | ||
docopt==0.6.2 | ||
docutils==0.14 | ||
dtaidistance==1.2.3 | ||
gensim==3.7.3 | ||
graphviz==0.9 | ||
idna==2.7 | ||
imbalanced-learn==0.5.0 | ||
imblearn==0.0 | ||
implicit==0.4.0 | ||
Jinja2==2.10.1 | ||
jmespath==0.9.3 | ||
joblib==0.13.2 | ||
JPype1==0.6.3 | ||
jsonschema==2.6.0 | ||
kiwisolver==1.0.1 | ||
lightgbm==2.2.2 | ||
MarkupSafe==1.1.1 | ||
matplotlib==3.0.2 | ||
networkx==2.2 | ||
nltk==3.4.3 | ||
numexpr==2.6.2 | ||
numpy==1.18.1 | ||
opencv-python-headless==4.0.0.21 | ||
pandas==0.23.4 | ||
pandas-profiling==1.4.1 | ||
pandasql==0.7.3 | ||
passlib==1.7.1 | ||
patsy==0.5.1 | ||
pg8000==1.13.1 | ||
pmdarima==1.1.0 | ||
protobuf==3.7.0 | ||
protobuf3-to-dict==0.1.5 | ||
psycopg2-binary==2.7.5 | ||
py4j==0.10.8.1 | ||
pyarrow==0.11.1 | ||
pycparser==2.19 | ||
pydotplus==2.0.2 | ||
pymssql==2.1.4 | ||
PyMySQL==0.9.3 | ||
pyparsing==2.3.0 | ||
python-dateutil==2.7.5 | ||
pytz==2018.7 | ||
PyYAML==3.13 | ||
redis==2.10.6 | ||
requests==2.20.1 | ||
s3transfer==0.1.13 | ||
sagemaker==1.18.5 | ||
scikit-learn==0.21.3 | ||
scipy==1.2.0 | ||
seaborn==0.9.0 | ||
six==1.12.0 | ||
sklearn==0.0 | ||
smart-open==1.8.3 | ||
soupsieve==1.9.1 | ||
SQLAlchemy==1.2.15 | ||
statsmodels==0.9.0 | ||
texttable==0.9.1 | ||
tqdm==4.28.1 | ||
twkorean==0.1.5 | ||
urllib3==1.24.1 | ||
websocket-client==0.55.0 | ||
xgboost==0.80 | ||
xlrd==1.2.0 |