Getting started: Training and Prediction with TensorFlow Estimator

Before using AI Platform with this tutorial, you should be familiar with machine learning and TensorFlow. To learn more, refer to Machine Learning Crash Course using TensorFlow APIs. For many more educational resources about machine learning, see Learn with Google AI.

Overview

This document provides an introductory, end-to-end walkthrough of training and prediction on AI Platform. You will walk through a sample that uses a census dataset to:

Create a TensorFlow 1.14 training application and validate it locally.
Run your training job on a single worker instance in the cloud.
Run your training job as a distributed training job in the cloud.
Optimize your hyperparameters by using hyperparameter tuning.
Deploy a model to support prediction.
Request an online prediction and see the response.
Request a batch prediction.

What you will build

The sample builds a wide and deep model for predicting income category based on United States Census Income Dataset. The two income categories (also known as labels) are:

>50K—Greater than 50,000 dollars
<=50K—Less than or equal to 50,000 dollars

Wide and deep models use deep neural nets (DNNs) to learn high-level abstractions about complex features or interactions between such features. These models then combine the outputs from the DNN with a linear regression performed on simpler features. This provides a balance between power and speed that is effective on many structured data problems.

You can read more about wide and deep models in the Google Research Blog post named Wide & Deep Learning: Better Together with TensorFlow.

The sample defines the model using TensorFlow's prebuilt DNNCombinedLinearClassifier class. The sample defines the data transformations particular to the census dataset, then assigns these (potentially) transformed features to either the DNN or the linear portion of the model.

Costs

This walkthrough uses billable components of Google Cloud, including:

AI Platform for:
- Training
- Prediction
Cloud Storage for:
- Storing input data for training
- Staging the training application package
- Writing training artifacts
- Storing input data files for batch prediction

Use the Pricing Calculator to generate a cost estimate based on your projected usage.

Set up and test your Cloud environment

Complete the following steps to set up a GCP account, activate the AI Platform API, and install and activate the Cloud SDK.

Set up your GCP project

Sign in to your Google Account.
If you don't already have one, sign up for a new account.
In the Cloud Console, on the project selector page, select or create a Cloud project.

Note: If you don't plan to keep the resources that you create in this procedure, create a project instead of selecting an existing project. After you finish these steps, you can delete the project, removing all resources associated with the project.

Go to the project selector page
Make sure that billing is enabled for your Google Cloud project. Learn how to confirm billing is enabled for your project.
Enable the AI Platform Training & Prediction and Compute Engine APIs.
Enable the APIs

Install and initialize the Cloud SDK.

Set up your environment

Choose one of the options below to set up your environment locally on macOS or in a remote environment on Cloud Shell.

For macOS users, we recommend that you set up your environment using the MACOS tab below. Cloud Shell, shown on the CLOUD SHELL tab, is available on macOS, Linux, and Windows. Cloud Shell provides a quick way to try AI Platform, but isn't suitable for ongoing development work.

macOS

Check Python installation
Confirm that you have Python installed and, if necessary, install it.
```
python -V
```
Check pip installation
pip is Python's package manager, included with current versions of Python. Check if you already have pip installed by running pip --version. If not, see how to install pip.

You can upgrade pip using the following command:
```
pip install -U pip
```
See the pip documentation for more details.
Install virtualenv
virtualenv is a tool to create isolated Python environments. Check if you already have virtualenv installed by running virtualenv --version. If not, install virtualenv:
```
pip install --user --upgrade virtualenv
```
To create an isolated development environment for this guide, create a new virtual environment in virtualenv. For example, the following command activates an environment named aip-env:
```
virtualenv aip-env
source aip-env/bin/activate
```
For the purposes of this tutorial, run the rest of the commands within your virtual environment.
See more information about using virtualenv. To exit virtualenv, run deactivate.

Cloud Shell

Open the Google Cloud Console.

Google Cloud Console
Click the Activate Google Cloud Shell button at the top of the console window.

A Cloud Shell session opens inside a new frame at the bottom of the console and displays a command-line prompt. It can take a few seconds for the shell session to be initialized.

Your Cloud Shell session is ready to use.
Configure the gcloud command-line tool to use your selected project.
```
gcloud config set project [selected-project-id]
```
where [selected-project-id] is your project ID. (Omit the enclosing brackets.)

Python version support

The sample for this tutorial uses Python 2.7. AI Platform Training and AI Platform Prediction also support Python 3. See how to select a Python version when submitting a training job.

Download the code for this tutorial

Download the sample from the GitHub repository.

macOS

Download and extract the AI Platform sample zip file.

Download the zip file
Open a terminal window and navigate to the directory that contains the extracted cloudml-samples-master directory.
Navigate to the cloudml-samples-master > census > estimator directory. The commands in this walkthrough must be run from the estimator directory.
```
cd cloudml-samples-master/census/estimator
```

Cloud Shell

Enter the following command to download the AI Platform sample zip file:

wget https://github.com/GoogleCloudPlatform/cloudml-samples/archive/master.zip

Unzip the file to extract the cloudml-samples-master directory.
```
unzip master.zip
```
Navigate to the cloudml-samples-master > census > estimator directory. The commands in this walkthrough must be run from the estimator directory.
```
cd cloudml-samples-master/census/estimator
```

Develop and validate your training application locally

Before you run your training application in the cloud, get it running locally. Local environments provide an efficient development and validation workflow so that you can iterate quickly. You also won't incur charges for cloud resources when debugging your application locally.

Get your training data

The relevant data files, adult.data and adult.test, are hosted in a public Cloud Storage bucket. For purposes of this sample, use the versions on Cloud Storage, which have undergone some trivial cleaning, instead of the original source data. See below for more information about the data.

You can read the data files directly from Cloud Storage or copy them to your local environment. For purposes of this sample you will download the samples for local training, and later upload them to your own Cloud Storage bucket for cloud training.

Download the data to a local file directory and set variables that point to the downloaded data files.
```
mkdir data
gsutil -m cp gs://cloud-samples-data/ai-platform/census/data/* data/
```
Set the TRAIN_DATA AND EVAL_DATA variables to your local file paths. For example, the following commands set the variables to local paths.
```
TRAIN_DATA=$(pwd)/data/adult.data.csv
EVAL_DATA=$(pwd)/data/adult.test.csv
```

The data is stored in comma-separated value format as shown by the following preview of the adult.data file:

39, State-gov, 77516, Bachelors, 13, Never-married, Adm-clerical, Not-in-family, White, Male, 2174, 0, 40, United-States, <=50K
50, Self-emp-not-inc, 83311, Bachelors, 13, Married-civ-spouse, Exec-managerial, Husband, White, Male, 0, 0, 13, United-States, <=50K
38, Private, 215646, HS-grad, 9, Divorced, Handlers-cleaners, Not-in-family, White, Male, 0, 0, 40, United-States, <=50K
53, Private, 234721, 11th, 7, Married-civ-spouse, Handlers-cleaners, Husband, Black, Male, 0, 0, 40, United-States, <=50K
...

Install dependencies

macOS

With the virtual environment that you activated in a previous step activated, run the following command:

pip install tensorflow==1.14.*

Running this command installs TensorFlow 1.14, which is used in the tutorial.

Cloud Shell

Although TensorFlow is installed on Cloud Shell, you must run the following command to ensure you are using the same version of TensorFlow required by the sample:

pip install --user tensorflow==1.14.*

Running this command installs TensorFlow 1.14, which is used in the tutorial.

Run a local training job

A local training job loads your Python training program and starts a training process in an environment that's similar to that of a live AI Platform cloud training job.

Specify an output directory and set a MODEL_DIR variable. The following command sets MODEL_DIR to a value of output.
```
MODEL_DIR=output
```
It's a good practice to delete the contents of the output directory in case data remains from a previous training run. The following command deletes all data in the output directory.

Caution: Be careful when using rm -rf. Check the variable name, and use rm -i if you want the system to ask you before deleting each file.
```
rm -rf $MODEL_DIR/*
```

To run your training locally, run the following command:

gcloud ai-platform local train \
    --module-name trainer.task \
    --package-path trainer/ \
    --job-dir $MODEL_DIR \
    -- \
    --train-files $TRAIN_DATA \
    --eval-files $EVAL_DATA \
    --train-steps 1000 \
    --eval-steps 100

By default, verbose logging is turned off. You can enable it by setting the --verbosity tag to DEBUG. A later example shows you how to enable it.

Inspect the summary logs using Tensorboard

To see the evaluation results, you can use the visualization tool called TensorBoard. With TensorBoard, you can visualize your TensorFlow graph, plot quantitative metrics about the execution of your graph, and show additional data like images that pass through the graph. Tensorboard is available as part of the TensorFlow installation.

Follow the steps below to launch TensorBoard and point it at the summary logs produced during training, both during and after execution.

macOS

Launch TensorBoard:
```
tensorboard --logdir=$MODEL_DIR
```
When you have started running TensorBoard, you can access it in your browser at http://localhost:6006

Cloud Shell

Launch TensorBoard:

tensorboard --logdir=$MODEL_DIR --port=8080

Select "Preview on port 8080" from the Web Preview menu at the top of the command line.

Click on Accuracy to see graphical representations of how accuracy changes as your job progresses.

Tensorboard accuracy graphs

You can shut down TensorBoard at any time by typing ctrl+c on the command line.

Run a local training job in distributed mode

You can test whether your model works with the AI Platform's distributed execution environment by running a local training job using the --distributed flag.

Specify an output directory and set MODEL_DIR variable again. The following command sets MODEL_DIR to a value of output-dist.
```
MODEL_DIR=output-dist
```
Delete the contents of the output directory in case data remains from a previous training run.

Caution: Be careful when using rm -rf. Check the variable name, and use rm -i if you want the system to ask you before deleting each file.
```
rm -rf $MODEL_DIR/*
```

Run the local train command using the --distributed option. Be sure to place the flag above the -- that separates the user arguments from the command-line arguments.

gcloud ai-platform local train \
    --module-name trainer.task \
    --package-path trainer/ \
    --job-dir $MODEL_DIR \
    --distributed \
    -- \
    --train-files $TRAIN_DATA \
    --eval-files $EVAL_DATA \
    --train-steps 1000 \
    --eval-steps 100

Inspect the output

Output files are written to the directory specified by --job-dir, which was set to output-dist:

ls -R output-dist/

You should see output similar to this:

checkpoint
eval
events.out.tfevents.1488577094.<host-name>
export
graph.pbtxt
model.ckpt-1000.data-00000-of-00001
model.ckpt-1000.index
model.ckpt-1000.meta
model.ckpt-2.data-00000-of-00001
model.ckpt-2.index
model.ckpt-2.meta

output-dist//eval:
events.out.tfevents.<timestamp>.<host-name>
events.out.tfevents.<timestamp><host-name>
events.out.tfevents.<timestamp>.<host-name>

output-dist//export:
census

output-dist//export/census:
<timestamp>

output-dist//export/census/<timestamp>:
saved_model.pb
variables
...

Inspect the logs

Inspect the summary logs using Tensorboard the same way that you did for the single-instance training job except that you must change the --logdir value to match the output directory name you used for distributed mode.

macOS

Launch TensorBoard:
```
tensorboard --logdir=$MODEL_DIR
```
When you have started running TensorBoard, you can access it in your browser at http://localhost:6006

Cloud Shell

Launch TensorBoard:

tensorboard --logdir=$MODEL_DIR --port=8080

Select "Preview on port 8080" from the Web Preview menu at the top of the command line.

Set up your Cloud Storage bucket

This section shows you how to create a new bucket. You can use an existing bucket, but it must be in the same region where you plan on running AI Platform jobs. Additionally, if it is not part of the project you are using to run AI Platform, you must explicitly grant access to the AI Platform service accounts.

Specify a name for your new bucket. The name must be unique across all buckets in Cloud Storage.
```
BUCKET_NAME="your_bucket_name"
```
For example, use your project name with -aiplatform appended:
```
PROJECT_ID=$(gcloud config list project --format "value(core.project)")
BUCKET_NAME=${PROJECT_ID}-aiplatform
```
Check the bucket name that you created.
```
echo $BUCKET_NAME
```
Select a region for your bucket and set a REGION environment variable.

Use the same region where you plan on running AI Platform jobs. See the available regions for AI Platform services.

For example, the following code creates REGION and sets it to us-central1:
```
REGION=us-central1
```
Create the new bucket:
```
gsutil mb -l $REGION gs://$BUCKET_NAME
```

Upload the data files to your Cloud Storage bucket.

Use gsutil to copy the two files to your Cloud Storage bucket.
```
gsutil cp -r data gs://$BUCKET_NAME/data
```

Set the TRAIN_DATA and EVAL_DATA variables to point to the files.

TRAIN_DATA=gs://$BUCKET_NAME/data/adult.data.csv
EVAL_DATA=gs://$BUCKET_NAME/data/adult.test.csv

Use gsutil again to copy the JSON test file test.json to your Cloud Storage bucket.
```
gsutil cp ../test.json gs://$BUCKET_NAME/data/test.json
```
Set the TEST_JSON variable to point to that file.
```
TEST_JSON=gs://$BUCKET_NAME/data/test.json
```

Run a single-instance training job in the cloud

With a validated training job that runs in both single-instance and distributed mode, you're now ready to run a training job in the cloud. You'll start by requesting a single-instance training job.

Use the default BASIC scale tier to run a single-instance training job. The initial job request can take a few minutes to start, but subsequent jobs run more quickly. This enables quick iteration as you develop and validate your training job.

Select a name for the initial training run that distinguishes it from any subsequent training runs. For example, you can append a number to represent the iteration.
```
JOB_NAME=census_single_1
```
Specify a directory for output generated by AI Platform by setting an OUTPUT_PATH variable to include when requesting training and prediction jobs. The OUTPUT_PATH represents the fully qualified Cloud Storage location for model checkpoints, summaries, and exports. You can use the BUCKET_NAME variable you defined in a previous step.

It's a good practice to use the job name as the output directory. For example, the following OUTPUT_PATH points to a directory named census_single_1.
```
OUTPUT_PATH=gs://$BUCKET_NAME/$JOB_NAME
```

Run the following command to submit a training job in the cloud that uses a single process. This time, set the --verbosity tag to DEBUG so that you can inspect the full logging output and retrieve accuracy, loss, and other metrics. The output also contains a number of other warning messages that you can ignore for the purposes of this sample.

gcloud ai-platform jobs submit training $JOB_NAME \
    --job-dir $OUTPUT_PATH \
    --runtime-version 1.14 \
    --python-version 2.7 \
    --module-name trainer.task \
    --package-path trainer/ \
    --region $REGION \
    -- \
    --train-files $TRAIN_DATA \
    --eval-files $EVAL_DATA \
    --train-steps 1000 \
    --eval-steps 100 \
    --verbosity DEBUG

You can monitor the progress of your training job by watching the command-line output or in AI Platform > Jobs on Google Cloud Console.

Inspect the output

In cloud training, outputs are produced into Cloud Storage. In this sample, outputs are saved to OUTPUT_PATH; to list them, run:

gsutil ls -r $OUTPUT_PATH

The outputs should be similar to the outputs from training locally (above).

Inspect the Cloud Logging logs

Logs are a useful way to understand the behavior of your training code on the cloud. When AI Platform runs a training job, it captures all stdout and stderr streams and logging statements. These logs are stored in Cloud Logging; they are visible both during and after execution.

The easiest way to find the logs for your job is to select your job in AI Platform > Jobs on Cloud Console, and then click "View logs".

If you leave "All logs" selected, you see all logs from all workers. You can also select a specific task; master-replica-0 gives you an overview of the job's execution from the master's perspective.

Because you selected verbose logging, you can inspect the full logging output. Look for the term accuracy in the logs for the AI Platform job.

If you want to view these logs in your terminal, you can do so from the command line with:

gcloud ai-platform jobs stream-logs $JOB_NAME

See all the options for gcloud ai-platform jobs stream-logs.

Inspect the summary logs using Tensorboard

You can inspect the behavior of your training job by launching TensorBoard and pointing it at the summary logs produced during training — both during and after execution.

Because the training programs write summaries directly to a Cloud Storage location, Tensorboard can automatically read from them without manual copying of event files.

macOS

Launch TensorBoard:
```
tensorboard --logdir=$OUTPUT_PATH
```
When you have started running TensorBoard, you can access it in your browser at http://localhost:6006

Cloud Shell

Launch TensorBoard:

tensorboard --logdir=$OUTPUT_PATH --port=8080

Select "Preview on port 8080" from the Web Preview menu at the top of the command line.

Click on Accuracy to see graphical representations of how accuracy changes as your job progresses.

You can shut down TensorBoard at any time by typing ctrl+c on the command line.

Run distributed training in the cloud

To take advantage of Google's scalable infrastructure when running training jobs, configure your training job to run in distributed mode.

No code changes are necessary to run this model as a distributed process in AI Platform.

To run a distributed job, set --scale-tier to any tier above basic. For more information about scale tiers, see the scale tier documentation.

Select a name for your distributed training job that distinguishes it from other training jobs. For example, you could use dist to represent distributed and a number to represent the iteration.
```
JOB_NAME=census_dist_1
```
Specify OUTPUT_PATH to include the job name so that you don't inadvertently reuse checkpoints between jobs. You might have to redefine BUCKET_NAME if you've started a new command-line session since you last defined it. For example, the following OUTPUT_PATH points to a directory named census-dist-1.
```
OUTPUT_PATH=gs://$BUCKET_NAME/$JOB_NAME
```

Run the following command to submit a training job in the cloud that uses multiple workers. Note that the job can take a few minutes to start.

Place --scale-tier above the -- that separates the user arguments from the command-line arguments. For example, the following command uses a scale tier of STANDARD_1:

gcloud ai-platform jobs submit training $JOB_NAME \
    --job-dir $OUTPUT_PATH \
    --runtime-version 1.14 \
    --python-version 2.7 \
    --module-name trainer.task \
    --package-path trainer/ \
    --region $REGION \
    --scale-tier STANDARD_1 \
    -- \
    --train-files $TRAIN_DATA \
    --eval-files $EVAL_DATA \
    --train-steps 1000 \
    --verbosity DEBUG  \
    --eval-steps 100

You can monitor the progress of your job by watching the command-line output or in AI Platform > Jobs on Google Cloud Console.

Inspect the logs

Inspect the Cloud Logging logs and summary logs the same way that you did for the single-instance training job.

For Cloud Logging logs: Either select your job in AI Platform > Jobs on Cloud Console, and then click View logs or use the following command from your terminal:

gcloud ai-platform jobs stream-logs $JOB_NAME

For TensorBoard:

macOS

Launch TensorBoard:
```
tensorboard --logdir=$OUTPUT_PATH
```
When you have started running TensorBoard, you can access it in your browser a http://localhost:6006

Cloud Shell

Launch TensorBoard:

tensorboard --logdir=$OUTPUT_PATH --port=8080

Select "Preview on port 8080" from the Web Preview menu at the top of the command line.

Hyperparameter Tuning

AI Platform offers hyperparameter tuning to help you maximize your model's predictive accuracy. The census sample stores the hyperparameter configuration settings in a YAML file named hptuning_config.yaml and includes the file in the training request using the --config variable.

Select a new job name and create a variable that references the configuration file.

HPTUNING_CONFIG=../hptuning_config.yaml
JOB_NAME=census_core_hptune_1
TRAIN_DATA=gs://$BUCKET_NAME/data/adult.data.csv
EVAL_DATA=gs://$BUCKET_NAME/data/adult.test.csv

Specify OUTPUT_PATH to include the job name so that you don't inadvertently reuse checkpoints between jobs. You might have to redefine BUCKET_NAME if you've started a new command-line session since you last defined it. For example, the following OUTPUT_PATH points to a directory named census_core_hptune_1.
```
OUTPUT_PATH=gs://$BUCKET_NAME/$JOB_NAME
```

Run the following command to submit a training job that not only uses multiple workers but also uses hyperparameter tuning.

gcloud ai-platform jobs submit training $JOB_NAME \
    --stream-logs \
    --job-dir $OUTPUT_PATH \
    --runtime-version 1.14 \
    --python-version 2.7 \
    --config $HPTUNING_CONFIG \
    --module-name trainer.task \
    --package-path trainer/ \
    --region $REGION \
    --scale-tier STANDARD_1 \
    -- \
    --train-files $TRAIN_DATA \
    --eval-files $EVAL_DATA \
    --train-steps 1000 \
    --verbosity DEBUG  \
    --eval-steps 100

For more information about hyperparameter tuning, see the hyperparameter tuning overview.

Deploy a model to support prediction

Choose a name for your model; this must start with a letter and contain only letters, numbers, and underscores. For example:
```
MODEL_NAME=census
```

Create an AI Platform model:

gcloud ai-platform models create $MODEL_NAME --regions=$REGION

Select the job output to use. The following sample uses the job named census_dist_1.
```
OUTPUT_PATH=gs://$BUCKET_NAME/census_dist_1
```
Look up the full path of your exported trained model binaries:
```
gsutil ls -r $OUTPUT_PATH/export
```
Find a directory named $OUTPUT_PATH/export/census/<timestamp> and copy this directory path (without the : at the end) and set the environment variable MODEL_BINARIES to its value. For example:
```
MODEL_BINARIES=gs://$BUCKET_NAME/census_dist_1/export/census/1487877383942/
```
Where $BUCKET_NAME is your Cloud Storage bucket name, and census_dist_1 is the output directory.

Run the following command to create a version v1:

gcloud ai-platform versions create v1 \
    --model $MODEL_NAME \
    --origin $MODEL_BINARIES \
    --runtime-version 1.14 \
    --python-version 2.7

You can get a list of your models using the models list command.

gcloud ai-platform models list

Send an online prediction request to a deployed model

You can now send prediction requests to your model. For example, the following command sends an online prediction request using a test.json file that you downloaded as part of the sample GitHub repository.

gcloud ai-platform predict \
    --model $MODEL_NAME \
    --version v1 \
    --json-instances ../test.json

The response includes the probabilities of each label (>50K and <=50K) based on the data entry in test.json, thus indicating whether the predicted income is greater than or less than 50,000 dollars.

The response looks like this:

CLASSES       PROBABILITIES
[u'0', u'1']  [0.9969545602798462, 0.0030454816296696663]

Submit a batch prediction job

The batch prediction service is useful if you have large amounts of data and no latency requirements on receiving prediction results. This uses the same format as online prediction, but requires data be stored in Cloud Storage.

Set a name for the job.
```
JOB_NAME=census_prediction_1
```

Set the output path.

OUTPUT_PATH=gs://$BUCKET_NAME/$JOB_NAME

Submit the prediction job.

gcloud ai-platform jobs submit prediction $JOB_NAME \
    --model $MODEL_NAME \
    --version v1 \
    --data-format text \
    --region $REGION \
    --input-paths $TEST_JSON \
    --output-path $OUTPUT_PATH/predictions

Unlike the previous commands, this one returns immediately. Check the progress of the job and wait for it to finish:

gcloud ai-platform jobs describe $JOB_NAME

You should see state: SUCCEEDED once the job completes; this may take several minutes. You can also see the job logs in your terminal using

gcloud ai-platform jobs stream-logs $JOB_NAME

Alternatively, you can check the progress in AI Platform > Jobs on Cloud Console.

After the job succeeds, you can:

Read the output summary.

gsutil cat $OUTPUT_PATH/predictions/prediction.results-00000-of-00001

You should see output similar to the following.

{"probabilities": [0.9962924122810364, 0.003707568161189556], "logits": [-5.593664646148682], "classes": 0, "logistic": [0.003707568161189556]}

List the other files that the job produced using the gsutil ls command.
```
gsutil ls -r $OUTPUT_PATH
```

Compared to online prediction, batch prediction:

Is slower for this small number of instances (but is more suitable for large numbers of instances).
May return output in a different order than the input (but the numeric index allows each output to be matched to its corresponding input instance; this is not necessary for online prediction since the outputs are returned in the same order as the original input instances).

After the predictions are available, the next step is usually to ingest these predictions into a database or data processing pipeline.

In this sample, you deployed the model before running the batch prediction, but it's possible to skip that step by specifying the model binaries URI when you submit the batch prediction job. One advantage of generating predictions from a model before deploying it is that you can evaluate the model's performance on different evaluation datasets to help you decide whether the model meets your criteria for deployment.

Cleaning up

If you've finished analyzing the output from your training and prediction runs, you can avoid incurring additional charges to your Google Cloud account for the Cloud Storage directories used in this guide:

Open a terminal window (if not already open).
Use the gsutil rm command with the -r flag to delete the directory that contains your most recent job:
```
gsutil rm -r gs://$BUCKET_NAME/$JOB_NAME
```

If successful, the command returns a message similar to this:

Removing gs://my-awesome-bucket/just-a-folder/cloud-storage.logo.png#1456530077282000...
Removing gs://my-awesome-bucket/...

Repeat the command for any other directories that you created for this sample.

Alternatively, if you have no other data stored in the bucket, you can run the gsutil rm -r command on the bucket itself.

About the data

The Census Income Data Set that this sample uses for training is hosted by the UC Irvine Machine Learning Repository.

Census data courtesy of Lichman, M. (2013). UCI Machine Learning Repository http://archive.ics.uci.edu/ml. Irvine, CA: University of California, School of Information and Computer Science. This dataset is publicly available for anyone to use under the following terms provided by the Dataset Source - http://archive.ics.uci.edu/ml - and is provided "AS IS" without any warranty, express or implied, from Google. Google disclaims all liability for any damages, direct or indirect, resulting from the use of the dataset.

What's next

You've now completed a walkthrough of an AI Platform sample that uses census data for training and prediction. You validated the training job locally, ran it in the cloud in both single-instance and distributed mode, used hyperparameter tuning to improve the model, and used the model to get online and batch predictions.

The following resources can help you continue learning about AI Platform.

Learn the basics of AI Platform.
See how to use GPUs and Cloud TPU nodes to train your models on AI Platform.
Follow an example that trains and predicts using low-level TensorFlow.