Training ResNet on Cloud TPU

The model in this tutorial is based on Deep Residual Learning for Image Recognition, which first introduces the residual network (ResNet) architecture. The tutorial uses the 50-layer variant, ResNet-50, and demonstrates training the model using TPUEstimator. The ResNet-50 model is pre-installed on your Compute Engine VM.


  • Create a Cloud Storage bucket to hold your dataset and model output.
  • Prepare a test version of the ImageNet dataset, referred to as the fake_imagenet dataset.
  • Run the training job.
  • Verify the output results.


This tutorial uses billable components of Google Cloud, including:

  • Compute Engine
  • Cloud TPU
  • Cloud Storage

Use the pricing calculator to generate a cost estimate based on your projected usage. 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 Account.

    If you don't already have one, sign up for a new account.

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

    Go to the project selector page

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

  4. 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.

  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 where you want to create 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 --project $PROJECT_ID

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

  5. Create a Cloud Storage bucket using the following command:

    gsutil mb -p ${PROJECT_ID} -c standard -l europe-west4 -b on gs://bucket-name

    This Cloud Storage bucket stores the data you use to train your model and the training results. The ctpu up tool used in this tutorial 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 required for this using the ctpu up command.

    ctpu up --project=${PROJECT_ID} \
    --zone=europe-west4-a \
    --vm-only \
    --name=resnet-tutorial \
    --disk-size-gb=300 \
    --machine-type=n1-standard-8 \

    Command flag descriptions

    Your GCP project ID
    The zone where you plan to create your Cloud TPU.
    Create a VM only. By default the ctpu up command creates a VM and a Cloud TPU.
    The name of the Cloud TPU to create.
    The size of the hard disk in GB of the VM created by the ctpu up command.
    The machine type of the Compute Engine VM to create.
    The version of Tensorflow ctpu installs on the VM.

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

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

When the ctpu up command has finished executing, verify that your shell prompt has changed from username@project to username@vm-name. This change shows that 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 resnet-tutorial --zone=europe-west4-a

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

Configure storage, model, and data paths

Set up the following environment variables, replacing bucket-name with the name of your Cloud Storage bucket:

(vm)$ export STORAGE_BUCKET=gs://bucket-name
(vm)$ export MODEL_DIR=${STORAGE_BUCKET}/resnet
(vm)$ export DATA_DIR=gs://cloud-tpu-test-datasets/fake_imagenet
(vm)$ export PYTHONPATH="${PYTHONPATH}:/usr/share/tpu/models"

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

Train and evaluate the ResNet model with fake_imagenet

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

This tutorial uses a demonstration version of the full ImageNet dataset, referred to as fake_imagenet. This demonstration version allows you to test the tutorial, while reducing the storage and time requirements typically associated with running a model against the full ImageNet database.

The fake_imagenet dataset is at this location on Cloud Storage:


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 will not be meaningful.

For information on how to download and process the full ImageNet dataset, see Downloading, preprocessing, and uploading the ImageNet dataset.

  1. Launch a Cloud TPU resource using the ctpu utility and set some environment variables used later on.

     (vm)$ ctpu up --project=${PROJECT_ID} \
       --tpu-only \
       --tf-version=1.15.5 \
     (vm)$ export TPU_NAME=resnet-tutorial
     (vm)$ export ACCELERATOR_TYPE=v3-8
  2. Navigate to the model directory:

    (vm)$ cd /usr/share/tpu/models/official/resnet/
  3. Run the training script.

    For a single Cloud TPU device, the script trains the ResNet-50 model for 90 epochs and evaluates the results after each training step. The number of training steps is set with the train_steps flag. Using the script command line below, the model should train in about 15 minutes.

    Since the training and evaluation is done on the fake_imagenet dataset, the training and evaluation results do not reflect the results that would be generated if training and evaluation was performed on a real dataset.

    If you run this script on a real dataset, use the train_steps flag to specify the number of training steps. See the .yaml files in the /usr/share/tpu/models/official/resnet/configs/cloud directory to get an idea about how many training steps to use.

     (vm)$ python3 \
        --tpu=${TPU_NAME} \
        --data_dir=${DATA_DIR} \
        --model_dir=${MODEL_DIR} \
        --train_steps=500 \
    Parameter Description
    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).
    data_dir Specifies the Cloud Storage path for training input. It is set to the fake_imagenet dataset in this example.
    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.
    config_file Specifies the YAML configuration file to use during training. The name of this file corresponds to the type of TPU used. For example, v2-8.yaml.

At this point, you can either conclude this tutorial and clean up your GCP resources, or you can further explore running the model on Cloud TPU Pods.

Scaling your model with Cloud TPU Pods

You can get results faster by scaling your model with Cloud TPU Pods. The fully supported ResNet-50 model can work with the following Pod slices:

  • v2-32
  • v2-128
  • v2-256
  • v2-512
  • v3-32
  • v3-128
  • v3-256
  • v3-512
  • v3-1024
  • v3-2048

When working with Cloud TPU Pods, you first train the model using a Pod, then use a single Cloud TPU device to evaluate the model.

Training with Cloud TPU Pods

  1. Delete the Cloud TPU resource you created for training the model on a single device.

     (vm)$ ctpu delete --project=${PROJECT_ID} \
       --tpu-only \
  2. Run the ctpu up command, using the tpu-size parameter to specify the Pod slice you want to use. For example, the following command uses a v2-32 Pod slice.

      (vm)$ ctpu up --project=${PROJECT_ID} \
        --tpu-only \
        --tpu-size=v2-32 \
        --tf-version=1.15.5 \
  3. Update the TPU_NAME and ACCELERATOR_TYPE environment variables to specify a TPU pod name an accelerator type.

      (vm)$ export TPU_NAME=resnet-tutorial
      (vm)$ export ACCELERATOR_TYPE=v2-32
  4. Update the MODEL_DIR directory to store the training data.

      (vm)$ export MODEL_DIR=${STORAGE_BUCKET}/resnet-tutorial
  5. Train the model, updating the config_file parameter to use the configuration file that corresponds with the Pod slice you want to use. For example, the training script uses the v2-32.yaml configuration file.

    The script trains the model on the fake_imagnet dataset to 35 epochs. This takes approximately 90 minutes to run on a v3-128 Cloud TPU.

      (vm)$ python3 \
        --tpu=${TPU_NAME} \
        --data_dir=${DATA_DIR} \
        --model_dir=${MODEL_DIR} \
        --train_steps=500 \
        --mode=train \

Evaluating the model

In this step, you use Cloud TPU to evaluate the above trained model against the fake_imagenet validation data.

  1. Delete the Cloud TPU resource you created to train the model.

     (vm)$ ctpu delete --project=${PROJECT_ID} \
       --tpu-only \
  2. Start a v2-8 Cloud TPU.

     (vm)$ ctpu up --project=${PROJECT_ID} \
       --tpu-only \
       --tf-version=1.15.5 \
  3. Update the TPU_NAME environment variable.

     (vm)$ export TPU_NAME=resnet-eval
  4. Run the model evaluation. This time, add the mode flag and set it to eval.

     (vm)$ python3 \
       --tpu=${TPU_NAME} \
       --data_dir=${DATA_DIR} \
       --model_dir=${MODEL_DIR} \
       --mode=eval \

This generates output similar to the following:

Eval results: {'loss': 8.255788, 'top_1_accuracy': 0.0009969076, 'global_step': 0, 'top_5_accuracy': 0.005126953}. Elapsed seconds: 76

Since the training and evaluation was done on the fake_imagenet dataset, the output results do not reflect actual output that would appear if the training and evaluation was performed on a real dataset.

Cleaning up

To avoid incurring charges to your Google Cloud account for the resources used in this tutorial, either delete the project that contains the resources, or keep the project and delete the individual resources.

  1. Disconnect from the Compute Engine instance, if you have not already done so:

    (vm)$ exit

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

  2. 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:

    $ ctpu delete --project=${PROJECT_ID} \
  3. 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:

    $ ctpu status --project=${PROJECT_ID}} \
     --name=resnet-tutorial \
    2018/04/28 16:16:23 WARNING: Setting zone to "europe-west4-a"
    No instances currently exist.
            Compute Engine VM:     --
            Cloud TPU:             --
  4. Run gsutil as shown, replacing bucket-name with the name of the Cloud Storage bucket you created for this tutorial:

    $ gsutil rm -r gs://bucket-name

What's next

In this tutorial you have trained the RESNET 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. Models trained on Cloud TPUs require datasets to be in TFRecord format.

You can use the dataset conversion tool sample 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 format yourself. For more information, see TFRecord and tf.Example

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. Information about model-specific hyperparameters can be found in the source code for each model. For more information on hyperparameter tuning, see Overview of hyperparameter tuning, Using the Hyperparameter tuning service and Tune hyperparameters.


Once you have trained your model you can use it for inference (also called prediction). AI Platform is a cloud-based solution for developing, training, and deploying machine learning models. Once a model is deployed, you can use the AI Platform Prediction service.

  • Learn more about ctpu, including how to install it on a local machine.
  • Explore the TPU tools in TensorBoard.
  • See how to train ResNet with Cloud TPU and GKE.
  • Walk through the tutorial for the RetinaNet object detection model.
  • Run the TensorFlow SqueezeNet model on Cloud TPU, using the above instructions as your starting point. The model architectures for SqueezeNet and ResNet-50 are similar. You can use the same data and the same command-line flags to train the model.