Training ResNet on Cloud TPU (TF 2.x)

---

This tutorial shows you how to train a Keras ResNet model on Cloud TPU using tf.distribute.TPUStrategy.

If you are not familiar with Cloud TPU, it is strongly recommended that you go through the quickstart for your framework to learn how to create a TPU and a Compute Engine VM.

Warning: 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.

Objectives

• Create a Cloud Storage bucket to hold your dataset and model output.
• Prepare a fake imagenet dataset that is similar to the ImageNet dataset.
• Run the training job.
• Verify the output results.

Costs

In this document, you use 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 Google Cloud 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 Google Cloud 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.

Cloud TPU single device training

Note: Do you want to train the model using a Cloud TPU pod? See TPU Pod training.

This section provides information on setting up Cloud Storage bucket, VM, and Cloud TPU resources for single device training.

Important: Set up all resources (Compute Engine VM, Cloud TPU, and Cloud Storage bucket) in the same region and zone to reduce network latency and network costs. VMs and TPU nodes are located in specific zones, which are subdivisions within a region.

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 the Google Cloud CLI 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 API calls with your credentials.

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

   Service accounts allow the Cloud TPU service to access other Google Cloud services.

   $ 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:

   $ gcloud storage buckets create gs://bucket-name --project=${PROJECT_ID} --location=us-central2

   This Cloud Storage bucket stores the data you use to train your model and the training results. The gcloud command used in this tutorial to set up the TPU also sets up default permissions for the Cloud TPU Service Account you set up in the previous step. If you want finer-grain permissions, review the access level permissions.

6. Prepare your dataset or use fake_imagenet

   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 fake_imagenet. This demonstration version lets you 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:

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

   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 won't be meaningful.

   If you want to use the full ImageNet dataset, see Downloading, preprocessing, and uploading the ImageNet dataset.

   Caution: For this tutorial, make sure you do not 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.

7. Launch TPU resources using the gcloud command. The command you use depends on whether you are using TPU VMs or TPU nodes. For more information on the two VM architecture, see System Architecture.

   Note: The following sample training uses the TF-Vision modeling library for computer vision. See the README for more information about the TF-Vision modeling library.

   $ gcloud compute tpus tpu-vm create resnet-tutorial \
     --zone=us-central2-b \
     --accelerator-type=v4-8 \
     --version=tpu-vm-tf-2.18.0-pjrt
   

   Command flag descriptions

   zone

   The zone where you plan to create your Cloud TPU.

   accelerator-type

   The accelerator type specifies the version and size of the Cloud TPU you want to create. For more information about supported accelerator types for each TPU version, see TPU versions.

   version

   The Cloud TPU software version.

   For more information on the gcloud command, see the gcloud Reference.

   Note: The first time you run gcloud compute tpus tpu-vm on a project it takes about 5 minutes to perform startup tasks such as SSH key propagation and API turnup.

8. Connect to the TPU VM instance using SSH. When you are connected to the VM, your shell prompt changes from username@projectname to username@vm-name:

   gcloud compute tpus tpu-vm ssh resnet-tutorial --zone=us-central2-b
   

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

9. Set the Cloud TPU name variable.

   (vm)$ export TPU_NAME=local
   

10. Set Cloud Storage bucket variables

    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=${STORAGE_BUCKET}/resnet-2x
    (vm)$ export DATA_DIR=gs://cloud-tpu-test-datasets/fake_imagenet
    
    

    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.

11. When creating your TPU, if you set the --version parameter to a version ending with -pjrt, set the following environment variables to enable the PJRT runtime:

      (vm)$ export NEXT_PLUGGABLE_DEVICE_USE_C_API=true
      (vm)$ export TF_PLUGGABLE_DEVICE_LIBRARY_PATH=/lib/libtpu.so
    

12. Install TensorFlow requirements.

    (vm)$ pip3 install -r /usr/share/tpu/models/official/requirements.txt 

13. The ResNet training script requires an extra package. Install it now:

    (vm)$ pip3 install --user tensorflow-model-optimization>=0.1.3
    

14. Change to directory that stores the model:

    (vm)$ cd /usr/share/tpu/tensorflow/resnet50_keras
    

15. Set the PYTHONPATH environment variable:

    (vm)$ /usr/share/tpu/tensorflow/resnet50_keras"
    

16. Run the training script. This uses a fake_imagenet dataset and trains ResNet for 100 steps.

    (vm)$ resnet50.py --tpu=local --data=gs://cloud-tpu-test-datasets/fake_imagenet
    

    Command flag descriptions

    tpu

    The name of your TPU.

    data

    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.

This will train ResNet for 100 steps and will complete on a v3-8 TPU node in approximately 3 minutes. At the end of the 100 steps, output similar to the following appears:

I0624 17:04:26.974905 140457742666816 controller.py:290]  eval | step:    100 | eval time:   23.3 sec | output:
    {'accuracy': 0.0010141226,
     'top_5_accuracy': 0.0051457332,
     'validation_loss': 8.448798}
 eval | step:    100 | eval time:   23.3 sec | output:
    {'accuracy': 0.0010141226,
     'top_5_accuracy': 0.0051457332,
     'validation_loss': 8.448798}

You have now completed the single-device training example. Use the following steps to delete the current single-device TPU resources.

1. Disconnect from the Compute Engine instance:

   (vm)$ exit
   

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

2. Delete the TPU resource.

   $ gcloud compute tpus tpu-vm delete resnet-tutorial \
     --zone=us-central2-b
   

   Command flag descriptions

   zone

   The zone where your Cloud TPU resided.

Scaling your model with Cloud TPU Pods

Training your model on Cloud TPU Pods may require some changes to your training script. For information, see Training on TPU Pods.

TPU Pod training

Important: If you have already set up a Cloud TPU project, service account, storage bucket, and dataset for single device training, you can skip to create TPU resources.

This section provides information on setting up a Cloud Storage bucket and Cloud TPU resources for Pod training.

Set up

Important: Set up your Cloud TPU resources and your Cloud Storage bucket in the same region and zone to reduce network latency and network costs. Cloud TPUs are located in specific zones, which are subdivisions within a region.

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 Google Cloud CLI 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 Google Cloud 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 or use a bucket you created earlier for your project:

   gcloud storage buckets create gs://bucket-name --project=${PROJECT_ID} --location=us-central2
   

   This Cloud Storage bucket stores the data you use to train your model and the training results. The gcloud command used in this tutorial sets up default permissions for the Cloud TPU Service Account you set up in the previous step. If you want finer-grain permissions, review the access level permissions.

   The bucket location must be in the same region as your TPU VM.

6. Prepare your dataset or use fake_imagenet

   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.

   The default Pod training accesses a demonstration version of the full ImageNet dataset, referred to as fake_imagenet. This demonstration version lets you test Pod training, while reducing the storage and time requirements typically associated with training a model against the full ImageNet database.

   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 won't be meaningful.

   If you want to use the full ImageNet dataset, see Downloading, preprocessing, and uploading the ImageNet dataset.

   Caution: Make sure you do not 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.

Create Cloud TPU resources

1. Launch your Cloud TPU resources using the gcloud command.

   This tutorial specifies a v3-32 Pod. For other Pod options, see TPU versions.

   Note: If there is not enough capacity currently available to create the TPU Pod, you can queue your request using queued resources. Queued resources allow you to receive capacity once it becomes available. To request your Cloud TPU resources as queued resources, use the gcloud compute tpus queued-resources create command instead. For more information, see Manage Queued Resources.

   $ gcloud compute tpus tpu-vm create resnet-tutorial \
     --zone=us-central2-b \
     --accelerator-type=4-32 \
     --version=tpu-vm-tf-2.18.0-pod-pjrt
   

   Command flag descriptions

   zone

   The zone where you plan to create your Cloud TPU.

   accelerator-type

   The accelerator type specifies the version and size of the Cloud TPU you want to create. For more information about supported accelerator types for each TPU version, see TPU versions.

   version

   The Cloud TPU software version.

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

2. Connect to the TPU VM instance using SSH. When you are connected to the VM, your shell prompt changes from username@projectname to username@vm-name:

   gcloud compute tpus tpu-vm ssh resnet-tutorial --zone=us-central2-b
   

   Key Point: From this point on, a prefix of (vm) $ means you should run the command on the Compute Engine VM instance. begins with (vm)$ in your VM session window.

3. Export Cloud TPU setup variables:

   (vm)$ export ZONE=us-central2-b
   (vm)$ export STORAGE_BUCKET=gs://bucket-name
   (vm)$ export TPU_NAME=resnet-tutorial
   (vm)$ export DATA_DIR=gs://cloud-tpu-test-datasets/fake_imagenet
   (vm)$ export MODEL_DIR=${STORAGE_BUCKET}/resnet-2x-pod
   

   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.

4. The ResNet training script requires an extra package. Install it now:

   (vm)$ pip3 install --user tensorflow-model-optimization>=0.1.3 

5. Install TensorFlow requirements. {: id='setup-env'}:

   (vm)$ pip3 install -r /usr/share/tpu/models/official/requirements.txt
   

6. Set the PYTHONPATH environment variable:

   (vm)$ export PYTHONPATH="PYTHONPATH=/usr/share/tpu/tensorflow/resnet50_keras"
   (vm)$ export TPU_LOAD_LIBRARY=0
   

7. Change to directory that stores the model:

   (vm)$ cd /usr/share/tpu/tensorflow/resnet50_keras
   

8. Train the model.

   (vm)$ resnet50.py --tpu=${TPU_NAME} --data=gs://cloud-tpu-test-datasets/fake_imagenet
   

   Command flag descriptions

   tpu

   The name of your TPU.

   data

   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 Cloud TPU of the same size and TensorFlow version.

This procedure trains the model on the fake_imagenet dataset to 100 training steps and 13 evaluation steps. This training takes approximately 2 minutes on a v3-32 Cloud TPU. When the training and evaluation complete, messages similar to the following appears:

{'accuracy': 0.0009716797,
     'learning_rate': 0.10256411,
     'top_5_accuracy': 0.0049560545,
     'training_loss': 8.5587225}
train | step:    100 | steps/sec:    1.2 | output:
    {'accuracy': 0.0009716797,
     'learning_rate': 0.10256411,
     'top_5_accuracy': 0.0049560545,
     'training_loss': 8.5587225}

eval | step:    100 | eval time:   24.8 sec | output:
    {'accuracy': 0.0010141226,
     'top_5_accuracy': 0.004356971,
     'validation_loss': 8.50038}
 eval | step:    100 | eval time:   24.8 sec | output:
    {'accuracy': 0.0010141226,
     'top_5_accuracy': 0.004356971,
     'validation_loss': 8.50038}

Clean 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@projectname, showing you are in the Cloud Shell.

2. Delete your Cloud TPU and Compute Engine resources.

   $ gcloud compute tpus tpu-vm delete resnet-tutorial \
     --zone=us-central2-b
   

3. Verify the resources have been deleted by running gcloud compute tpus tpu-vm list. The deletion might take several minutes. The output from the following command shouldn't include any of the TPU resources created in this tutorial:

   $ gcloud compute tpus tpu-vm list --zone=us-central2-b

4. Delete the Cloud Storage bucket you created for this tutorial:

   $ gcloud storage rm gs://bucket-name --recursive
   

Note: For free storage limits and other pricing information, see the Cloud Storage pricing guide.

What's next

The TensorFlow Cloud TPU tutorials generally train the model using a sample dataset. The results of this training are not usable for inference. To use a model for inference, you can train the data on a publicly available dataset or your own dataset. TensorFlow models trained on Cloud TPUs generally 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 and Tune hyperparameters.

Inference

Once you have trained your model, you can use it for inference (also called prediction). You can use the Cloud TPU inference converter tool to prepare and optimize a TensorFlow model for inference on Cloud TPU v5e. For more information about inference on Cloud TPU v5e, see Cloud TPU v5e inference introduction.

• Learn how to train and evaluate using your own data in place of the fake_imagenet or ImageNet datasets by following the dataset conversion tutorial. The tutorial explains how to use the image classification data converter example script to convert a raw dataset for image classification into TFRecords usable by Cloud TPU TensorFlow models.
• Run a Cloud TPU colab that demonstrates how to run an image classification model using your own image data.
• Explore the other Cloud TPU tutorials.
• Learn to use the TPU monitoring tools in TensorBoard.
• See how to train ResNet with Cloud TPU and GKE.