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 to learn how to create a TPU and a Compute Engine VM.

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

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 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.

    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. 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 bucket, 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 tpu.googleapis.com --project $PROJECT_ID
    

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

    service-PROJECT_NUMBER@cloud-tpu.iam.gserviceaccount.com
    

  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 gcloud compute tpus execution-groups tool 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 Compute Engine (VM) and your Cloud TPU node.

  6. Launch a Compute Engine VM and Cloud TPU using the gcloud command.

    $ gcloud compute tpus execution-groups create \
     --vm-only \
     --name=resnet-tutorial \
     --zone=europe-west4-a \
     --disk-size=300 \
     --machine-type=n1-standard-16 \
     --tf-version=2.5.0
    

    Command flag descriptions

    vm-only
    Create a VM only. By default the gcloud compute tpus execution-groups command creates a VM and a Cloud TPU.
    name
    The name of the Cloud TPU to create.
    zone
    The zone where you plan to create your Cloud TPU.
    disk-size
    The size of the hard disk in GB of the VM created by the gcloud compute tpus execution-groups command.
    machine-type
    The machine type of the Compute Engine VM to create.
    tf-version
    The version of Tensorflow gcloud compute tpus execution-groups installs on the VM.

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

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

    When the gcloud compute tpus execution-groups command has finished executing, verify that your shell prompt has changed from username@projectname to username@vm-name. This change shows that you are now logged into your Compute Engine VM.

    gcloud compute ssh resnet-tutorial --zone=europe-west4-a
    

    As you continue these instructions, run each command that begins with (vm)$ in your Compute Engine instance.

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
(vm)$ export PYTHONPATH="$PYTHONPATH:/usr/share/models/"

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.

Train and evaluate the ResNet model with 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 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:

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 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 gcloud command.

    (vm)$ gcloud compute tpus execution-groups create \
     --tpu-only \
     --accelerator-type=v3-8  \
     --name=resnet-tutorial \
     --zone=europe-west4-a \
     --tf-version=2.5.0
    

    Command flag descriptions

    tpu-only
    Creates the Cloud TPU without creating a VM. By default the gcloud compute tpus execution-groups command creates a VM and a Cloud TPU.
    accelerator-type
    The type of the Cloud TPU to create.
    name
    The name of the Cloud TPU to create.
    zone
    The zone where you plan to create your Cloud TPU.
    tf-version
    The version of Tensorflow to install on the VM.
  2. Set the Cloud TPU name variable.

    (vm)$ export TPU_NAME=resnet-tutorial
    
  3. The ResNet training script requires an extra package. Install it now:

    (vm)$ sudo pip3 install tensorflow-model-optimization>=0.1.3
    
  4. Navigate to the ResNet-50 model directory:

    (vm)$ cd /usr/share/models/official/vision/image_classification/resnet/
    
  5. Run the training script. This uses a fake_imagenet dataset and trains ResNet for one epoch.

    (vm)$ python3 resnet_ctl_imagenet_main.py \
     --tpu=${TPU_NAME} \
     --model_dir=${MODEL_DIR} \
     --data_dir=${DATA_DIR} \
     --batch_size=1024 \
     --steps_per_loop=500 \
     --train_epochs=1 \
     --use_synthetic_data=false \
     --dtype=fp32 \
     --enable_eager=true \
     --enable_tensorboard=true \
     --distribution_strategy=tpu \
     --log_steps=50 \
     --single_l2_loss_op=true \
     --use_tf_function=true
    

    Command flag descriptions

    tpu
    The name of your Cloud TPU.
    model_dir
    The Cloud Storage bucket where checkpoints and summaries are stored during training. You can use an existing folder to load previously generated checkpoints created on a TPU of the same size and TensorFlow version.
    data_dir
    The Cloud Storage path of training input. It is set to the fake_imagenet dataset in this example.
    batch_size
    The training batch size.
    steps_per_loop
    The number of training steps to run before saving state to the CPU. A training step is the processing of one batch of examples. This includes both a forward pass and back propagation.
    train_epochs
    The number of epochs to train the model.
    use_synthetic_data
    Set to true to use synthetic data for training.
    dtype
    The data type to use for training.
    enable_eager
    Enable TensorFlow eager execution.
    enable_tensorboard
    Enable TensorBoard.
    distribution_strategy
    Set to tpu to train the ResNet model on a Cloud TPU.
    log_steps
    The number of training steps to take before logging timing information such as examples per second.
    single_l2_loss_op
    Set to true to calculate L2_loss on concatenated weights, instead of using Keras per-layer L2 loss.
    use_tf_function
    Set to true to wrap the train and test steps inside a tf.function.

This will train ResNet for 1 epoch and will complete on a v3-8 TPU node in under 10 minutes. The training script output should include text like:

{
  'train_loss': 1.435225,
  'train_accuracy': 0.00084427913
}

The training script also performs evaluation. The evaluation output should contain text like this:

Run stats:
{
  'eval_loss': 0.861013,
  'eval_acc': 0.001,
  'train_loss': 1.435225,
  'train_acc': 0.00084427913,
  'step_timestamp_log': [
    'BatchTimestamp<batch_index: 0,
    timestamp: 1606330585.7613473>',
    'BatchTimestamp<batch_index: 500,
    timestamp: 1606330883.8486104>',
    'BatchTimestamp<batch_index: 1000,
    timestamp: 1606331119.515312>',
    'BatchTimestamp<batch_index: 1251,
    timestamp: 1606331240.7516596>'
  ],
  'train_finish_time': 1606331296.395158,
  'avg_exp_per_second': 1951.6983246161021
}

To train the ResNet to convergence, run it for 90 epochs as shown in the following script. Training and evaluation are done together. Each epoch has 1251 steps for a total of 112590 training steps and 48 evaluation steps.

(vm)$ python3 resnet_ctl_imagenet_main.py \
    --tpu=${TPU_NAME} \
    --model_dir=${MODEL_DIR} \
    --data_dir=${DATA_DIR} \
    --batch_size=1024 \
    --steps_per_loop=500 \
    --train_epochs=90 \
    --use_synthetic_data=false \
    --dtype=fp32 \
    --enable_eager=true \
    --enable_tensorboard=true \
    --distribution_strategy=tpu \
    --log_steps=50 \
    --single_l2_loss_op=true \
    --use_tf_function=true

Command flag descriptions

tpu
The name of your Cloud TPU.
model_dir
The Cloud Storage bucket where checkpoints and summaries are stored during training. You can use an existing folder to load previously generated checkpoints created on a TPU of the same size and TensorFlow version.
data_dir
The Cloud Storage path of training input. It is set to the fake_imagenet dataset in this example.
batch_size
The training batch size.
steps_per_loop
The number of training steps to run before saving state to the CPU. A training step is the processing of one batch of examples. This includes both a forward pass and back propagation.
train_epochs
The number of epochs to train the model.
use_synthetic_data
Whether to use synthetic data for training.
dtype
The data type to use for training.
enable_eager
Enable TensorFlow eager execution.
enable_tensorboard
Enable TensorBoard.
distribution_strategy
Set to tpu to train the ResNet model on a Cloud TPU.
log_steps
The number of training steps to take before logging timing information such as examples per second.
single_l2_loss_op
Set to true to calculate L2_loss on concatenated weights, instead of using Keras per-layer L2 loss.
use_tf_function
Set to true to wrap the train and test steps inside a tf.function.

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.

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
  • v3-32

With Cloud TPU Pods, training and evaluation are done together.

Training with Cloud TPU Pods

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

    (vm)$ gcloud compute tpus execution-groups delete resnet-tutorial \
      --zone=europe-west4-a \
      --tpu-only
  2. Run the gcloud compute tpus execution-groups command, using the accelerator-type parameter to specify the Pod slice you want to use. For example, the following command uses a v3-32 Pod slice.

    (vm)$ gcloud compute tpus execution-groups  create --name=resnet-tutorial \
      --accelerator-type=v3-32  \
      --zone=europe-west4-a \
      --tf-version=2.5.0 \
      --tpu-only
    

    Command flag descriptions

    name
    The name of the Cloud TPU to create.
    accelerator-type
    The type of the Cloud TPU to create.
    zone
    The zone where you plan to create your Cloud TPU.
    tf-version
    The version of Tensorflow gcloud installs on the VM.
    tpu-only
    Create a Cloud TPU only. By default the gcloud command creates a VM and a Cloud TPU.
  3. Set some required environment variables:

    (vm)$ export MODEL_DIR=${STORAGE_BUCKET}/resnet-2x-pod
    
  4. Train the model.

    (vm)$ python3 resnet_ctl_imagenet_main.py \
      --tpu=${TPU_NAME} \
      --model_dir=${MODEL_DIR} \
      --data_dir=${DATA_DIR} \
      --batch_size=4096 \
      --steps_per_loop=500 \
      --train_epochs=1 \
      --use_synthetic_data=false \
      --dtype=fp32 \
      --enable_eager=true \
      --enable_tensorboard=true \
      --distribution_strategy=tpu \
      --log_steps=50 \
      --single_l2_loss_op=true \
      --use_tf_function=true

    Command flag descriptions

    tpu
    The name of your Cloud TPU.
    model_dir
    The Cloud Storage bucket where checkpoints and summaries are stored during training. You can use an existing folder to load previously generated checkpoints created on a TPU of the same size and TensorFlow version.
    data_dir
    The Cloud Storage path of training input. It is set to the fake_imagenet dataset in this example.
    batch_size
    The training batch size.
    steps_per_loop
    The number of training steps to run before saving state to the CPU. A training step is the processing of one batch of examples. This includes both a forward pass and back propagation.
    train_epochs
    The number of epochs to train the model.
    use_synthetic_data
    Set to true to use synthetic data for training.
    dtype
    The data type to use for training.
    enable_eager
    Set to true to enable TensorFlow eager execution.
    enable_tensorboard
    Set to true to enable TensorBoard.
    distribution_strategy
    Set to tpu to train the ResNet model on a TPU.
    log_steps
    The number of training steps to take before logging timing information such as examples per second.
    single_l2_loss_op
    Calculate L2_loss on concatenated weights, instead of using Keras per-layer L2 loss.
    use_tf_function
    Wrap the train and test steps inside a tf.function.

The procedure trains the model on the fake_imagenet dataset to 1 epoch (312 total training steps and 12 evaluation steps). This training takes approximately 2 minutes on a v3-32 Cloud TPU. The training script should display text like the following:

step: 312
steps_per_second: 1.38
{
  'train_loss': 0.3755003,
  'train_accuracy': 0.000983605
}

The script also performs evaluation and should display text like the following:

step: 312
evaluation metric: {
  'test_loss': 0.20280758,
  'test_accuracy': 0.001
}

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@projectname, showing you are in the Cloud Shell.

  2. In your Cloud Shell, use the following command to delete your Compute Engine VM and Cloud TPU:

    $ gcloud compute tpus execution-groups delete resnet-tutorial \
      --zone=europe-west4-a
    
  3. Verify the resources have been deleted by running gcloud compute tpus execution-groups list. The deletion might take several minutes. A response like the one below indicates your instances have been successfully deleted.

    $ gcloud compute tpus execution-groups list \
     --zone=europe-west4-a
    

    You should see an empty list of TPUs like the following:

       NAME             STATUS
    
  4. Delete your Cloud Storage bucket using gsutil as shown below. Replace bucket-name with the name of your Cloud Storage bucket.

    $ 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.

Inference

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 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.