Create README.md

samples/tfx-oss/README.md

@@ -0,0 +1,53 @@
+# TFX Pipeline Example
+
+This sample walks through running [TFX](https://github.com/tensorflow/tfx) [Taxi Application Example](https://github.com/tensorflow/tfx/tree/master/examples/chicago_taxi_pipeline) on Kubeflow Pipelines cluster. 
+
+## Overview
+
+This pipeline demonstrates the TFX capablities at scale. The pipeline uses a public BigQuery dataset and uses GCP services to preprpocess data (Dataflow) and train the model (Cloud ML Engine). The model is then deployed to Cloud ML Engine Prediction service.
+
+
+## Setup
+
+Create a local Python 3.5 conda environment
+```
+conda create -n tfx-kfp pip python=3.5.3
+```
+then activate the environment.
+
+
+Install TFX and Kubeflow Pipelines SDK
+```
+!pip3 install https://storage.googleapis.com/ml-pipeline/tfx/tfx-0.12.0rc0-py2.py3-none-any.whl 
+!pip3 install https://storage.googleapis.com/ml-pipeline/release/0.1.10/kfp.tar.gz --upgrade
+```
+
+Clone TFX github repo
+```
+git clone https://github.com/tensorflow/tfx
+```
+
+Upload the utility code to your storage bucket. You can modify this code if needed for a different dataset.
+```
+gsutil cp tfx/examples/chicago_taxi_pipeline/taxi_utils.py gs://my-bucket/
+```
+
+## Configure the TFX Pipeline
+
+Modify the pipeline configuration file at 
+```
+tfx/examples/chicago_taxi_pipeline/taxi_pipeline_kubeflow_large.py
+```
+Configure 
+- GCS storage bucket name (replace "my-bucket")
+- GCP project ID (replace "my-gcp-project")
+- Make sure the path to the taxi_utils.py is correct
+- Set the limit on the BigQuery query. The original dataset has 100M rows, which can take time to process. Set it to 20000 to run an sample test.
+
+
+## Compile a run the pipeline
+```
+python tfx/examples/chicago_taxi_pipeline/taxi_pipeline_kubeflow_large.py
+```
+This will generate a file named chicago_taxi_pipeline_kubeflow_large.tar.gz 
+Upload this file to the Pipelines Cluster and crate a run.