Running Inception on Cloud TPU

This tutorial shows you how to train the Inception model on Cloud TPU.

Disclaimer

This tutorial uses a third-party dataset. Google provides no representation, warranty, or other guarantees about the validity, or any other aspects of, this dataset.

Warning: This tutorial uses the TPUEstimator API which is only supported by TensorFlow 1.x. If you are writing a model with TensorFlow 2.x, use Keras instead.

Model description

Inception v3 is a widely-used image recognition model that can attain significant accuracy. The model is the culmination of many ideas developed by multiple researchers over the years. It is based on the original paper: "Rethinking the Inception Architecture for Computer Vision" by Szegedy, et. al.

The model has a mixture of symmetric and asymmetric building blocks, including:

• convolutions
• average pooling
• max pooling
• concatenate
• dropouts
• fully connected layers

Loss is computed using Softmax.

The following picture shows the model at a high-level:

You can find more information about the model at GitHub.

The model is built using the high-level Estimator API.

The API greatly simplifies model creation by encapsulating most low-level functions, allowing users to focus on model development, not the inner workings of the underlying hardware that runs things.

Objectives

• Create a Cloud Storage bucket to hold your dataset and model output.
• Run the training job.
• Verify the output results.

Costs

This tutorial uses the following billable components of Google Cloud:

• Compute Engine
• Cloud TPU
• Cloud Storage

To generate a cost estimate based on your projected usage, use the pricing calculator. New Google Cloud users might be eligible for a free trial.

Before you begin

Before starting this tutorial, check that your Google Cloud project is correctly set up.

1. Sign in to your Google Cloud account. If you're new to Google Cloud, create an account to evaluate how our products perform in real-world scenarios. New customers also get $300 in free credits to run, test, and deploy workloads.

2. In the Google Cloud Console, on the project selector page, select or create a Google 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 project selector

3. Make sure that billing is enabled for your Cloud project. Learn how to confirm that billing is enabled for your project.

4. In the Google Cloud Console, on the project selector page, select or create a Google 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 project selector

5. Make sure that billing is enabled for your Cloud project. Learn how to confirm that billing is enabled for your project.

This walkthrough uses billable components of Google Cloud. Check the Cloud TPU pricing page to estimate your costs. Be sure to clean up resources you create when you've finished with them to avoid unnecessary charges.

Set up your resources

This section provides information on setting up Cloud Storage, VM, and Cloud TPU resources for tutorials.

Important: Set up all resources in the same region/zone to reduce network latency and network costs.

1. Open a Cloud Shell window.

   Open Cloud Shell

2. Create a variable for your project's ID.

   export PROJECT_ID=project-id
   

3. Configure gcloud command-line tool to use the project in which you want to create your Cloud TPU.

   gcloud config set project ${PROJECT_ID}
   

   The first time you run this command in a new Cloud Shell VM, an Authorize Cloud Shell page is displayed. Click Authorize at the bottom of the page to allow gcloud to make GCP API calls with your credentials.

4. Create a Service Account for the Cloud TPU project.

   gcloud beta services identity create --service tpu.googleapis.com --project $PROJECT_ID
   

   The command returns a Cloud TPU Service Account with following format:

   [e-mail]
   

5. Create a Cloud Storage bucket using the following command. Replace bucket-name with a name for your bucket.

   gsutil mb -p ${PROJECT_ID} -c standard -l us-central1 -b on gs://bucket-name
   

   The Cloud Storage bucket stores the data you use to train your model and the training results. The ctpu up tool sets up default permissions for the Cloud TPU Service Account. If you want finer-grain permissions, review the access level permissions.

   The bucket location must be in the same region as your virtual machine (VM) and your TPU node. VMs and TPU nodes are located in specific zones, which are subdivisions within a region.

6. Launch the Compute Engine resources using the ctpu up command.

   Note: This tutorial uses TensorFlow 1.15.5 which is based on the Debian 10 image and defaults to Python 3.6. Earlier versions of TensorFlow 1.15.x were based on Debian 9 and default to Python 3.5.

   ctpu up --project=${PROJECT_ID} \
    --zone=us-central1-b \
    --vm-only \
    --machine-type=n1-standard-8 \
    --tf-version=1.15.5 \
    --name=inception-tutorial
   

   Command flag descriptions

   project

   Your GCP project ID

   zone

   The zone where you plan to create your Cloud TPU.

   vm-only

   Creates the VM without creating a Cloud TPU. By default the ctpu up command creates a VM and a Cloud TPU.

   machine-type

   The machine type of the Compute Engine VM to create.

   tf-version

   The version of TensorFlow ctpu installs on the VM.

   name

   The name of the Cloud TPU to create.

   For more information on the CTPU utility, see CTPU Reference.

7. When prompted, press y to create your Cloud TPU resources.

   Note: The first time you run ctpu up on a project it takes about 5 minutes to perform startup tasks such as SSH key propagation and API turnup.

   When the ctpu up command has finished executing, verify that your shell prompt has changed from username@projectname to username@vm-name. This verifies you are now logged into your Compute Engine VM. If you are not connected to the Compute Engine instance, you can do so by running the following command:

   gcloud compute ssh inception-tutorial --zone=us-central1-b
   

   From this point on, a prefix of (vm)$ means you should run the command on the Compute Engine VM instance.

8. Create an environment variable for the storage bucket. Replace bucket-name with the name of your Cloud Storage bucket.

   (vm)$ export STORAGE_BUCKET=gs://bucket-name
   

9. Create an environment variable for the TPU name.

   (vm)$ export TPU_NAME=inception-tutorial

Training dataset

The training application expects your training data to be accessible in Cloud Storage. The training application also uses your Cloud Storage bucket to store checkpoints during training.

ImageNet is an image database. The images in the database are organized into a hierarchy, with each node of the hierarchy depicted by hundreds and thousands of images.

This tutorial uses a demonstration version of the full ImageNet dataset, referred to as the fake_imagenet dataset. This dataset enables you to test the tutorial, without requiring the storage or time that required to download and run a model against the full ImageNet database. Alternatively, you can use the full ImageNet dataset.

A DATA_DIR environment variable is used to specify the dataset on which to train.

Note that the fake_imagenet dataset is only useful for understanding how to use a Cloud TPU, and validating end-to-end performance. The accuracy numbers and saved model are not meaningful.

The fake_imagenet dataset is the following location on Cloud Storage:

gs://cloud-tpu-test-datasets/fake_imagenet

Note: Make sure you don't set the STORAGE_BUCKET environment variable to the path of the fake_imagenet dataset. You can read from gs://cloud-tpu-test-datasets but you can't write to it. As a result, you can't use it to write out training logs. Make sure the STORAGE_BUCKET environment variable is set to your own Cloud Storage bucket.

(Optional) Set up TensorBoard

TensorBoard offers a suite of tools designed to present TensorFlow data visually. When used for monitoring, TensorBoard can help identify bottlenecks in processing and suggest ways to improve performance.

If you don't need to monitor the model's output at this time, you can skip the TensorBoard setup steps.

If you want to monitor the model's output and performance, follow the guide to setting up TensorBoard.

Run the model

You are now ready to train and evaluate the Inception v3 model using ImageNet data.

The Inception v3 model is pre-installed on your Compute Engine VM, in the /usr/share/tpu/models/experimental/inception/ directory.

In the following steps, a prefix of (vm)$ means you should run the command on your Compute Engine VM:

1. Set up a DATA_DIR environment variable containing one of the following values:

   • If you are using the fake_imagenet dataset:

     (vm)$ export DATA_DIR=gs://cloud-tpu-test-datasets/fake_imagenet
     

   • If you have uploaded a set of training data to your Cloud Storage bucket:

     (vm)$ export DATA_DIR=${STORAGE_BUCKET}/data
     

2. Run the Inception v3 model:

   (vm)$ python /usr/share/tpu/models/experimental/inception/inception_v3.py \
       --tpu=$TPU_NAME \
       --learning_rate=0.165 \
       --train_steps=250000 \
       --iterations=500 \
       --use_tpu=True \
       --use_data=real \
       --mode=train_and_eval \
       --train_steps_per_eval=2000 \
       --data_dir=${DATA_DIR} \
       --model_dir=${STORAGE_BUCKET}/inception

   • --tpu specifies the name of the Cloud TPU. Note that ctpu passes this name to the Compute Engine VM as an environment variable (TPU_NAME).
   • --use_data specifies which type of data the program must use during training, either fake or real. The default value is fake.
   • --data_dir specifies the Cloud Storage path for training input. The application ignores this parameter when you're using fake_imagenet data.
   • --model_dir specifies the directory where checkpoints and summaries are stored during model training. If the folder is missing, the program creates one. When using a Cloud TPU, the model_dir must be a Cloud Storage path (gs://...). You can reuse an existing folder to load current checkpoint data and to store additional checkpoints as long as the previous checkpoints were created using TPU of the same size and TensorFlow version.

What to expect

Inception v3 operates on 299x299 images. The default training batch size is 1024, which means that each iteration operates on 1024 of those images.

You can use the --mode flag to select one of three modes of operation: train, eval, and train_and_eval:

• --mode=train or --mode=eval specifies either a training-only or an evaluation-only job.
• --mode=train_and_eval specifies a hybrid job that does both training and evaluation.

Train-only jobs run for the specified number of steps defined in train_steps and can go through the entire training set, if desired.

Train_and_eval jobs cycle though training and evaluation segments. Each training cycle runs for train_steps_per_eval and is followed by an evaluation job (using the weights that have been trained up to that point).

The number of training cycles is defined by the floor function of train_steps divided by train_steps_per_eval.

    floor(train_steps / train_steps_per_eval)

Note: The train_steps_per_eval flag must be a multiple of iterations.

By default, Estimator API-based models report loss values every certain number of steps. The reporting format is along the lines of: step = 15440, loss = 12.6237

Discussion: TPU-specific modifications to the model

The specific modifications required to get Estimator API-based models ready for TPUs are surprisingly minimal. The program imports the following libraries: from google.third_party.tensorflow.contrib.tpu.python.tpu import tpu_config from google.third_party.tensorflow.contrib.tpu.python.tpu import tpu_estimator from google.third_party.tensorflow.contrib.tpu.python.tpu import tpu_optimizer The CrossShardOptimizer function wraps the optimizer, as in: if FLAGS.use_tpu: optimizer = tpu_optimizer.CrossShardOptimizer(optimizer) The function that defines the model returns an Estimator specification using: return tpu_estimator.TPUEstimatorSpec( mode=mode, loss=loss, train_op=train_op, eval_metrics=eval_metrics) The main function defines an Estimator-compatible configuration using: run_config = tpu_config.RunConfig( master=tpu_grpc_url, evaluation_master=tpu_grpc_url, model_dir=FLAGS.model_dir, save_checkpoints_secs=FLAGS.save_checkpoints_secs, save_summary_steps=FLAGS.save_summary_steps, session_config=tf.ConfigProto( allow_soft_placement=True, log_device_placement=FLAGS.log_device_placement), tpu_config=tpu_config.TPUConfig( iterations_per_loop=iterations, num_shards=FLAGS.num_shards, per_host_input_for_training=per_host_input_for_training)) The program uses this defined configuration and a model definition function to create an Estimator object: inception_classifier = tpu_estimator.TPUEstimator( model_fn=inception_model_fn, use_tpu=FLAGS.use_tpu, config=run_config, params=params, train_batch_size=FLAGS.train_batch_size, eval_batch_size=eval_batch_size, batch_axis=(batch_axis, 0)) Train-only jobs need only to call the train function: inception_classifier.train( input_fn=imagenet_train.input_fn, steps=FLAGS.train_steps) Evaluation-only jobs get their data from available checkpoints and wait until a new one becomes available: for checkpoint in get_next_checkpoint(): eval_results = inception_classifier.evaluate( input_fn=imagenet_eval.input_fn, steps=eval_steps, hooks=eval_hooks, checkpoint_path=checkpoint) When you choose the option train_and_eval, the training and the evaluation jobs run in parallel. During evaluation, trainable variables are loaded from the latest available checkpoint. Training and evaluation cycles repeat as you specify in the flags:: ``` for cycle in range(FLAGS.train_steps // FLAGS.train_steps_per_eval): inception_classifier.train( input_fn=imagenet_train.input_fn, steps=FLAGS.train_steps_per_eval)

  eval_results = inception_classifier.evaluate(
      input_fn=imagenet_eval.input_fn, steps=eval_steps, hooks=eval_hooks)

If you used the fake\_imagenet dataset to train the model, proceed to
[clean up](#clean-up).

## Using the full Imagenet dataset {: #full-dataset }

The ImageNet dataset consists of three parts, training data, validation data,
and image labels.

The training data contains 1000 categories and 1.2 million images,  packaged for
easy downloading. The validation and test data are not contained in the ImageNet
training data (duplicates have been removed).

The validation and test data consists of 150,000 photographs, collected from
[Flickr](https://www.flickr.com/) and other search engines, hand labeled with
the presence or absence of 1000 object categories. The 1000 object categories
contain both internal nodes and leaf nodes of ImageNet, but do not overlap with
each other. A random subset of 50,000 of the images with labels has been
released as validation data along with a list of the 1000 categories. The
remaining images are used for evaluation and have been released without labels.

### Steps to pre-processing the full ImageNet dataset

There are five steps to preparing the full ImageNet dataset for use by a Machine
Learning model:

1. Verify that you have space on the download target.
1. Set up the target directories.
1. Register on the ImageNet site and request download permission.
1. Download the dataset to local disk or Compute Engine VM.

   Note: Downloading the Imagenet dataset to a Compute Engine VM takes
   considerably longer than downloading to your local machine (approximately 40
   hours versus 7 hours). However, if you download the dataset to your local
   machine, you must copy the files to a Compute Engine VM to pre-process them
   and then upload them to Cloud Storage before using them to train your model. Copying
   all of the training and validation files from your local machine to the VM
   takes about 13 hours. The recommended approach is to download the dataset to
   a VM.

1. Run the pre-processing and upload script.

### Verify space requirements

Whether you download the dataset to your local machine or to a Compute Engine
VM, you need about 300GB of space available on the download target. On a VM, you
can check your available storage with the `df -ha` command.

Note: If you use `gcloud compute` to set up your VM, it will allocate 250 GB by
default.

You can increase the size of the VM disk using one of the following methods:

*  Specify the `--disk-size` flag on the `gcloud compute` command line with the
   size, in GB, that you want allocated.
*  Follow the Compute Engine guide to [add a disk][add-disk] to your
   VM.
   * Set **When deleting instance** to **Delete disk** to ensure that the
     disk is removed when you remove the VM.
   * Make a note of the path to your new disk. For example: `/mnt/disks/mnt-dir`.

### Set up the target directories

On your local machine or Compute Engine VM, set up the directory structure to
store the downloaded data.

*  Create and export a home directory for the ImageNet dataset.

   Create a directory, for example, `imagenet` under your home directory on
   your local machine or VM. Under this directory, create two sub-directories:
   `train` and `validation`. Export the home directory as IMAGENET_HOME:

   <pre class="prettyprint lang-sh tat-dataset">
   export IMAGENET_HOME=~/imagenet
   </pre>

### Register and request permission to download the dataset

*  Register on the [Imagenet website](http://image-net.org/). You cannot
   download the dataset until ImageNet confirms your registration and sends you
   a confirmation email. If you do not get the confirmation email within a
   couple of days, contact [ImageNet support](mailto:[e-mail]) to
   see why your registration has not been confirmed. Once your registration is
   confirmed, you can download the dataset. The Cloud TPU tutorials that use the
   ImageNet dataset use the images from the ImageNet Large Scale Visual
   Recognition Challenge 2012 (ILSVRC2012).

### Download the ImageNet dataset

1. From the [LSRVC 2012 download site](https://image-net.org/challenges/LSVRC/2012/2012-downloads.php),
   go to the Images section on the page and right-click
   "Training images (Task 1 & 2)". This will give you the URL needed to download
   the largest part of the training set. Save the URL.

   Right-click "Training images (Task 3)" to get the URL for the second
   training set. Save the URL.

   Right-click "Validation images (all tasks)" to get the URL for the
   validation dataset. Save the URL.

   If you download the ImageNet files to your local machine, you need to copy
   the directories on your local machine to the corresponding `$IMAGENET_HOME`
   directory on your Compute Engine VM. Copying the ImageNet dataset from
   local host to your VM takes approximately 13 hours.

   For example, the following command copies all of the files under
   $IMAGENET_HOME on your local machine to your VM that displays the shell
   prompt <var>username@vm-name</var>:

   <pre class="prettyprint lang-sh tat-dataset">
   gcloud compute scp --recurse $IMAGENET_HOME <var>username@vm-name</var>:~/imagenet
   </pre>

1. From $IMAGENET_HOME, use `wget` to download the training and validation files
   using the saved URLs.

   The "Training images (Task 1 & 2)" file is the large training set. It is
   138GB and if you are downloading to a Compute Engine VM using the Cloud
   Shell, the download takes approximately 40 hours. If the Cloud Shell loses its
   connection to the VM, you can prepend `nohup` to the command or use
   [screen](https://linuxize.com/post/how-to-use-linux-screen/).

   <pre class="prettyprint lang-sh tat-dataset">
   cd $IMAGENET_HOME \
   nohup wget http://image-net.org/challenges/LSVRC/2012/dd31405981ef5f776aa17412e1f0c112/ILSVRC2012_img_train.tar
   </pre>

   This command downloads a large tar file: ILSVRC2012_img_train.tar.

   From $IMAGENET_HOME on the VM, extract the individual training directories
   into the `$IMAGENET_HOME/train` directory using the following command. The
   extraction takes between 1 - 3 hours.

   <pre class="prettyprint lang-sh tat-dataset">
   tar xf ILSVRC2012_img_train.tar
   </pre>

   Extract the individual training tar files located in the $IMAGENET_HOME/train
   directory, as shown in the following script:

   <pre class="prettyprint lang-sh tat-dataset">
   cd $IMAGENET_HOME/train

   for f in *.tar; do
     d=`basename $f .tar`
     mkdir $d
     tar xf $f -C $d
   done
   </pre>

   After you have extracted all of the tar files you can delete them to free
   up disk space.

   The "Training images (Task 3)" file is 728 MB and takes just a few minutes
   to download so you do not need to take precautions against losing the Cloud
   Shell connection.

   When you download this file, it extracts the individual training directories
   into the existing `$IMAGENET_HOME/train` directory.

   <pre class="prettyprint lang-sh tat-dataset">
   wget http://www.image-net.org/challenges/LSVRC/2012/dd31405981ef5f776aa17412e1f0c112/ILSVRC2012_img_train_t3.tar
   </pre>

   The "Validation images (all tasks)" file is 6GB, so you might want to use
   `nohup` or [screen](https://linuxize.com/post/how-to-use-linux-screen/) to
   avoid having the download exit if the Cloud Shell disconnects.

   <pre class="prettyprint lang-sh tat-dataset">
   wget http://www.image-net.org/challenges/LSVRC/2012/dd31405981ef5f776aa17412e1f0c112/ILSVRC2012_img_val.tar
   </pre>

   This download takes about 30 minutes. When you download this file, it
   extracts the individual validation directories into the
   `$IMAGENET_HOME/validation` directory.

   If you downloaded the validation files to your local machine, you need to
   copy the `$IMAGENET_HOME/validation` directory on your local machine to the
   `$IMAGENET_HOME/validation` directory on your Compute Engine VM. This copy
   operation takes about 30 minutes.

   Download the labels file. This takes a few seconds.

   <pre class="prettyprint lang-sh tat-dataset">
   wget -O $IMAGENET_HOME/synset_labels.txt \
https://raw.githubusercontent.com/tensorflow/models/master/research/inception/inception/data/imagenet_2012_validation_synset_labels.txt
   </pre>

   If you downloaded the labels file to your local machine, you need to copy it
   to the `$IMAGENET_HOME` directory on your local machine to `$IMAGENET_HOME`
   on your Compute Engine VM. This copy operation takes a few seconds.

   The training subdirectory names (for example, n03062245) are "WordNet IDs"
   (wnid). The [ImageNet API](https://image-net.org/download-attributes.php)
   shows the mapping of WordNet IDs to their associated validation labels in the
   `synset_labels.txt` file. A synset in this context is a visually similar
   group of images.

### Process the Imagenet dataset and, optionally, upload to Cloud Storage

1. Download the `imagenet_to_gcs.py` script from GitHub:

   <pre class="prettyprint lang-sh tat-dataset">
   wget https://raw.githubusercontent.com/tensorflow/tpu/master/tools/datasets/imagenet_to_gcs.py
   </pre>

1. If you are uploading the dataset to Cloud Storage, specify the storage
   bucket location to upload the ImageNet dataset:

   <pre class="lang-sh prettyprint tat-client-exports">
   export STORAGE_BUCKET=gs://<var>bucket-name</var>
   </pre>

1. If you are uploading the dataset to your local machine or VM, specify a data
   directory to hold the dataset:

   <pre class="lang-sh prettyprint tat-client-exports">
   <span class="no-select">(vm)$ </span>export DATA_DIR=$IMAGENET_HOME/<var>dataset-directory</var>
   </pre>

1. Run the script to pre-process the raw dataset as TFRecords and upload it to
   Cloud Storage using the following command:

   Note: If you don't want to upload to Cloud Storage, specify `--nogcs_upload`
   as another parameter and leave off the `--project` and `--gcs_output_path`
   parameters.

   <pre class="prettypring lang-sh tat-dataset">
     python3 imagenet_to_gcs.py \
      --project=$PROJECT \
      --gcs_output_path=$STORAGE_BUCKET  \
      --raw_data_dir=$IMAGENET_HOME \
      --local_scratch_dir=$IMAGENET_HOME/tf_records
   </pre>

Note: Downloading and preprocessing the data can take 10 or more hours,
depending on your network and computer speed. Do not interrupt the script.

The script generates a set of directories (for both training and validation) of
the form:

    ${DATA_DIR}/train-00000-of-01024
    ${DATA_DIR}/train-00001-of-01024
     ...
    ${DATA_DIR}/train-01023-of-01024

and

    ${DATA_DIR}/validation-00000-of-00128
    S{DATA_DIR}/validation-00001-of-00128
     ...
    ${DATA_DIR}/validation-00127-of-00128

After the data has been uploaded to your Cloud bucket, run your model and set
`--data_dir=${DATA_DIR}`.








## Clean up {: #clean-up }

To avoid incurring charges to your GCP account for the resources used
in this topic:

1. Disconnect from the Compute Engine VM:

    <pre class="lang-sh prettyprint tat-skip">
    <span class="no-select">(vm)$ </span>exit
    </pre>

     Your prompt should now be `username@projectname`, showing you are in the
     Cloud Shell.

1. In your Cloud Shell, run `ctpu delete` with the --zone flag you used when
   you set up the Cloud TPU to delete your
   Compute Engine VM and your Cloud TPU:

    <pre class="lang-sh prettyprint tat-resource-setup">
    <span class="no-select">$ </span>ctpu delete [optional: --zone]
    </pre>

    Important: If you set the TPU resources name when you ran `ctpu up`, you must
    specify that name with the `--name` flag when you run `ctpu delete` in
    order to shut down your TPU resources.

1. Run `ctpu status` to make sure you have no instances allocated to avoid
   unnecessary charges for TPU usage. The deletion might take several minutes.
   A response like the one below indicates there are no more allocated
   instances:

    <pre class="lang-sh prettyprint tat-skip">
    <span class="no-select">$ </span>ctpu status --zone=europe-west4-a
    </pre>
    <pre class="lang-sh prettyprint tat-skip">
    2018/04/28 16:16:23 WARNING: Setting zone to "--zone=europe-west4-a"
    No instances currently exist.
        Compute Engine VM:     --
        Cloud TPU:             --
    </pre>

1. Run `gsutil` as shown, replacing <var>bucket-name</var> with the name of the
   Cloud Storage bucket you created for this tutorial:

    <pre class="lang-sh prettyprint tat-resource-setup">
    <span class="no-select">$ </span>gsutil rm -r gs://<var>bucket-name</var>
    </pre>

Note: For free storage limits and other pricing information, see the
[Cloud Storage pricing guide](/storage/pricing).


## Inception v4

The Inception v4 model is a deep neural network model that uses Inception v3
building blocks to achieve higher accuracy than Inception v3. It is described in
the paper "Inception-v4, Inception-ResNet and the Impact of Residual Connections
on Learning" by Szegedy et. al.

The Inception v4 model is pre-installed on your Compute Engine VM, in
the `/usr/share/tpu/models/experimental/inception/` directory.

In the following steps, a prefix of `(vm)$` means you should run the command on
your Compute Engine VM:

1. If you have TensorBoard running in your Cloud Shell tab, you need another tab
   to work in. Open another tab in your Cloud Shell, and use `ctpu` in the new
   shell to connect to your Compute Engine VM:

    <pre class="lang-sh prettyprint">
    <span class="no-select">$ </span>ctpu up --project=${PROJECT_ID} </pre>

1. Set up a `DATA_DIR` environment variable containing one of the following
   values:

    * If you are using the fake\_imagenet dataset:

        <pre class="prettyprint lang-sh">
        <span class="no-select">(vm)$ </span>export DATA_DIR=gs://cloud-tpu-test-datasets/fake_imagenet
        </pre>

    * If you have uploaded a set of training data to your Cloud Storage
      bucket:

        <pre class="prettyprint lang-sh">
        <span class="no-select">(vm)$ </span>export DATA_DIR=${STORAGE_BUCKET}/data
        </pre>

1. Run the Inception v4 model:

    <pre class="lang-sh prettyprint">
    <span class="no-select">(vm)$ </span>python /usr/share/tpu/models/experimental/inception/inception_v4.py \
        --tpu=$TPU_NAME \
        --learning_rate=0.36 \
        --train_steps=1000000 \
        --iterations=500 \
        --use_tpu=True \
        --use_data=real \
        --train_batch_size=256 \
        --mode=train_and_eval \
        --train_steps_per_eval=2000 \
        --data_dir=${DATA_DIR} \
        --model_dir=${STORAGE_BUCKET}/inception</pre>

    * `--tpu` specifies the name of the Cloud TPU. Note that `ctpu`
      passes this name to the Compute Engine VM as an environment
      variable (`TPU_NAME`).
    * `--use_data` specifies which type of data the program must use during
      training, either fake or real. The default value is fake.
    * `--train_batch_size` specifies the train batch size to be 256. As the
      Inception v4 model is larger than Inception v3, it must be run at a
      smaller batch size per TPU core.
    * `--data_dir` specifies the Cloud Storage path for training input.
      The application ignores this parameter when you're using fake\_imagenet
      data.
    * `--model_dir` specifies the directory where checkpoints and
      summaries are stored during model training. If the folder is missing, the
      program creates one. When using a Cloud TPU, the `model_dir`
      must be a Cloud Storage path (`gs://...`). You can reuse an existing
      folder to load current checkpoint data and to store additional
      checkpoints as long as the previous checkpoints were created using TPU of
      the same size and TensorFlow version.

    

Clean up

What's next

## What's next {: #whats-next } The TensorFlow Cloud TPU tutorials generally train the model using a sample dataset. The results of this training are (in most cases) not usable for inference. To use a model for inference you can train the data on a publicly available dataset or your own data set. TensorFlow models trained on Cloud TPUs generally require datasets to be in [TFRecord](https://www.tensorflow.org/tutorials/load_data/tfrecord) format. You can use the [dataset conversion tool sample](https://cloud.google.com/tpu/docs/classification-data-conversion) to convert an image classification dataset into TFRecord format. If you are not using an image classification model, you will have to convert your dataset to [TFRecord](https://www.tensorflow.org/tutorials/load_data/tfrecord) format yourself. For more information, see [TFRecord and tf.Example](https://www.tensorflow.org/tutorials/load_data/tfrecord). ### Hyperparameter tuning To improve the model's performance with your dataset, you can tune the model's hyperparameters. You can find information about hyperparameters common to all TPU supported models on [GitHub](https://github.com/tensorflow/tpu/tree/master/models/hyperparameters). Information about model-specific hyperparameters can be found in the [source code](https://github.com/tensorflow/tpu/tree/master/models/official) for each model. For more information on hyperparameter tuning, see [Overview of hyperparameter tuning](https://cloud.google.com/ai-platform/training/docs/hyperparameter-tuning-overview), [Using the Hyperparameter tuning service](https://cloud.google.com/ai-platform/training/docs/using-hyperparameter-tuning), and [Tune hyperparameters](https://developers.google.com/machine-learning/guides/text-classification/step-5). ### Inference Once you have trained your model you can use it for inference (also called prediction). [AI Platform](https://cloud.google.com/ai-platform/docs/technical-overview) is a cloud-based solution for developing, [training](https://cloud.google.com/ai-platform/training/docs), and [deploying](https://cloud.google.com/ai-platform/prediction/docs/deploying-models) machine learning models. Once a model is deployed, you can use the [AI Platform Prediction service](https://cloud.google.com/ai-platform/prediction/docs). * Go in depth with an [advanced view](/tpu/docs/inception-v3-advanced) of Inception v3 on Cloud TPU. * Learn more about [`ctpu`](https://github.com/tensorflow/tpu/tree/master/tools/ctpu), including how to install it on a local machine. * Explore the [TPU tools in TensorBoard](/tpu/docs/cloud-tpu-tools){: track-type="gettingStarted" track-name="tutorialLink" track-metadata-position="nextSteps"}.