cleaning up the kubeflow examples to explicitly mention installation …

…of s2i in the README, updating the rquirements files to include pandas in the data_download pipeline step, update the requirements to point to the tar of the en_core_web_sm package since it cannot be pip installed, and small formatting changes to pipeline steps (SeldonIO#3654)

Co-authored-by: Alex Lee <alexander.lee@nytimes.com>

examples/kubeflow/README.md

@@ -22,6 +22,7 @@ Make sure you have the following components set-up and running in your Kubernete
 
 * [Seldon Core installed](https://docs.seldon.io/projects/seldon-core/en/latest/workflow/install.html#install-seldon-core-with-helm) with an [ingress (Ambassador / Istio) set up](https://docs.seldon.io/projects/seldon-core/en/latest/workflow/install.html#ingress-support)
 * Kubeflow Pipelines [version 1.0.0 Standalone](https://www.kubeflow.org/docs/pipelines/installation/standalone-deployment/#deploying-kubeflow-pipelines) set up
+* [s2i](https://github.com/openshift/source-to-image#installation) installed on your machine
 
 Let's get started! 🚀🔥 We will be building the end-to-end pipeline below:
 

examples/kubeflow/pipeline/pipeline_steps/clean_text/Transformer.py

@@ -1,7 +1,8 @@
 import re
+import logging
 from html.parser import HTMLParser
+
 import numpy as np
-import logging
 
 
 class Transformer:

examples/kubeflow/pipeline/pipeline_steps/clean_text/pipeline_step.py

@@ -1,6 +1,6 @@
 import dill
 import click
-import dill
+
 try:
     # Running for tests
     from .Transformer import Transformer

examples/kubeflow/pipeline/pipeline_steps/clean_text/requirements.txt

@@ -1,4 +1,3 @@
 dill==0.3.2
 click==7.1.2
 numpy==1.19.1
-

examples/kubeflow/pipeline/pipeline_steps/data_downloader/pipeline_step.py

@@ -1,6 +1,7 @@
 import click
-import numpy as np
 import dill
+
+import numpy as np
 import pandas as pd
 
 @click.command()

examples/kubeflow/pipeline/pipeline_steps/data_downloader/requirements.txt

@@ -2,3 +2,4 @@ dill==0.3.2
 click==7.1.2
 numpy==1.19.1
 scikit-learn==0.23.2
+pandas==1.1.1

examples/kubeflow/pipeline/pipeline_steps/lr_text_classifier/pipeline_step.py

@@ -1,6 +1,7 @@
 import click
-import numpy as np
 import dill
+
+import numpy as np
 from sklearn.linear_model import LogisticRegression
 
 @click.command()

examples/kubeflow/pipeline/pipeline_steps/spacy_tokenize/requirements.txt

@@ -2,5 +2,4 @@ dill==0.3.2
 click==7.1.2
 numpy==1.19.1
 spacy==2.3.2
-en-core-web-sm==2.3.1
-
+https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-2.3.1/en_core_web_sm-2.3.1.tar.gz

examples/kubeflow/pipeline/pipeline_steps/tfidf_vectorizer/pipeline_step.py

@@ -1,6 +1,7 @@
 import click
-import numpy as np
 import dill
+
+import numpy as np
 from sklearn.feature_extraction.text import TfidfVectorizer
 
 @click.command()

examples/kubeflow/pipeline/pipeline_steps/tfidf_vectorizer/requirements.txt

@@ -2,4 +2,3 @@ dill==0.3.2
 click==7.1.2
 numpy==1.19.1
 scikit-learn==0.23.2
-

examples/kubeflow/requirements-dev.txt

@@ -1,3 +1,4 @@
+dill==0.3.2
 python-dateutil==2.8.1
 kfp==1.0.0
 kubernetes==11.0.0
@@ -6,5 +7,5 @@ seldon_core==1.2.3
 numpy==1.19.1
 pandas==1.1.1
 spacy==2.3.2
-scikit-learn-0.23.2
-en-core-web-sm=2.3.1
+scikit-learn==0.23.2
+https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-2.3.1/en_core_web_sm-2.3.1.tar.gz