PyTorch MNIST Tutorial

This tutorial describes how to port an existing PyTorch model to Determined. We will port a simple image classification model for the MNIST dataset. This tutorial is based on the official PyTorch MNIST example.


  • Access to a Determined cluster. If you have not yet installed Determined, refer to the installation instructions.

  • Access to the Determined CLI on your local machine. See the installation instructions if you do not already have it installed. After installing the CLI, configure it to connect to your Determined cluster by setting the DET_MASTER environment variable to the hostname or IP address where Determined is running.


To use a PyTorch model in Determined, you need to port the model to Determined’s API. For most models, this porting process is straightforward, and once the model has been ported, all of the features of Determined will then be available: for example, you can do distributed training or hyperparameter search without changing your model code, and Determined will store and visualize your model metrics automatically.

When training a PyTorch model, Determined provides a built-in training loop that feeds each batch of data into your train_batch function, where you will perform the forward pass, the backpropagation, and the computation of training metrics for that batch. Determined also handles checkpointing, log management, and device initialization. To plug your model code into the Determined training loop, you define methods to perform the following tasks:

  • initialize the models, optimizers, and LR schedulers

  • define the training function for forward and backward passes

  • load the training data set

  • load the validation data set

The Determined training loop will then invoke these functions automatically. These methods should be organized into a trial class, which is a user-defined Python class that inherits from determined.pytorch.PyTorchTrial. The following sections walk through how to write your first trial class and then how to run a training job with Determined.

The complete code for this tutorial can be downloaded here: mnist_pytorch.tgz. After downloading this file, open a terminal window, extract the file, and cd into the mnist_pytorch directory:

tar xzvf mnist_pytorch.tgz
cd mnist_pytorch

We suggest you follow along with the code as you read through this tutorial.

Building a PyTorchTrial Class

Here is what the skeleton of our trial class looks like:

import torch.nn as nn
from determined.pytorch import DataLoader, PyTorchTrial, PyTorchTrialContext

class MNISTTrial(PyTorchTrial):
    def __init__(self, context: PyTorchTrialContext):
        # Initialize the trial class and wrap the models, optimizers, and LR schedulers.

    def train_batch(self, batch: TorchData, epoch_idx: int, batch_idx: int):
        # Run the training forward passes on the models and backward passes on the optimizers

    def evaluate_batch(self, batch: TorchData):
        # Define how to evaluate the model by calculating loss and other metrics.
        # for a batch of validation data.

    def build_training_data_loader(self):
        # Create the training data loader.
        # This should return a determined.pytorch.Dataset.

    def build_validation_data_loader(self):
        # Create the validation data loader.
        # This should return a determined.pytorch.Dataset.

We now discuss how to implement each of these methods in more detail.


As with any Python class, the __init__ method is invoked to construct our trial class. Determined passes this method a single parameter, an instance of PyTorchTrialContext, which inherits from TrialContext. The trial context contains information about the trial, such as the values of the hyperparameters to use for training. All the models and optimizers must be wrapped with wrap_model and wrap_optimizer respectively provided by PyTorchTrialContext. This MNIST model code uses the Torch Sequential API and torch.optim.Adadelta. The current values of the model’s hyperparameters can be accessed via the get_hparam() method of the trial context.

def __init__(self, context: PyTorchTrialContext):
    # Store trial context for later use.
    self.context = context

    # Initialize the model and wrap it using self.context.wrap_model().
    self.model = self.context.wrap_model(
            nn.Conv2d(1, self.context.get_hparam("n_filters1"), 3, 1),
            nn.Linear(144 * self.context.get_hparam("n_filters2"), 128),
            nn.Linear(128, 10),

    # Initialize the optimizer and wrap it using self.context.wrap_optimizer().
    self.optimizer = self.context.wrap_optimizer(
            model.parameters(), lr=self.context.get_hparam("learning_rate")

Loading Data

The last two methods we need to define are build_training_data_loader and build_validation_data_loader. Determined uses these methods to load the training and validation datasets, respectively. Both methods should return a determined.pytorch.DataLoader, which is very similar to

def build_training_data_loader(self):
    if not self.data_downloaded:
        self.download_directory = data.download_dataset(
        self.data_downloaded = True

    train_data = data.get_dataset(self.download_directory, train=True)
    return DataLoader(train_data, batch_size=self.context.get_per_slot_batch_size())

def build_validation_data_loader(self):
    if not self.data_downloaded:
        self.download_directory = data.download_dataset(
        self.data_downloaded = True

    validation_data = data.get_dataset(self.download_directory, train=False)
    return DataLoader(
        validation_data, batch_size=self.context.get_per_slot_batch_size()

For more information on loading data in Determined, refer to the tutorial on Accessing Data.

Defining train_batch

The train_batch method runs the forward passes through the models and the backward passes on the losses and steps the optimizers with a single batch of data. Determined expects a dictionary with the user-defined metrics and will automatically average all the metrics across batches. If an optimizer is set to automatically handle zeroing out the gradients, step_optimizer will zero out the gradients and there will be no need to call optim.zero_grad().

def train_batch(self, batch: TorchData, epoch_idx: int, batch_idx: int):
    batch = cast(Tuple[torch.Tensor, torch.Tensor], batch)
    data, labels = batch

    # Define the training forward pass and calculate loss.
    output = self.model(data)
    loss = torch.nn.functional.nll_loss(output, labels)

    # Define the training backward pass and step the optimizer.

    return {"loss": loss}

def evaluate_batch(self, batch: TorchData):
    batch = cast(Tuple[torch.Tensor, torch.Tensor], batch)
    data, labels = batch

    output = self.model(data)
    validation_loss = torch.nn.functional.nll_loss(output, labels).item()

    pred = output.argmax(dim=1, keepdim=True)
    accuracy = pred.eq(labels.view_as(pred)).sum().item() / len(data)

    return {"validation_loss": validation_loss, "accuracy": accuracy}

Training the Model

Now that we have ported our model code to the trial API, we can use Determined to train a single instance of the model or to do a hyperparameter search. In Determined, a trial is a training task that consists of a dataset, a deep learning model, and values for all of the model’s hyperparameters. An experiment is a collection of one or more trials: an experiment can either train a single model (with a single trial), or can define a search over a user-defined hyperparameter space.

To create an experiment, we start by writing a configuration file that defines the kind of experiment we want to run. In this case, we want to train a single model for a single epoch, using fixed values for the model’s hyperparameters:

description: mnist_pytorch_const
  learning_rate: 1.0
  global_batch_size: 64
  n_filters1: 32
  n_filters2: 64
  dropout1: 0.25
  dropout2: 0.5
records_per_epoch: 50_000
  name: single
  metric: validation_loss
    epochs: 1
  smaller_is_better: true
entrypoint: model_def:MNistTrial

The entrypoint specifies the name of the trial class to use. This is useful if our model code contains more than one trial class. In this case, we use an entrypoint of model_def:MNistTrial because our trial class is named MNistTrial and it is defined in a Python file named

For more information on experiment configuration, see the experiment configuration reference.

Running an Experiment

The Determined CLI can be used to create a new experiment, which will immediately start running on the cluster. To do this, we run:

det experiment create const.yaml .

Here, the first argument (const.yaml) is the name of the experiment configuration file and the second argument (.) is the location of the directory that contains our model definition files. You may need to configure the CLI with the network address where the Determined master is running, via the -m flag or the DET_MASTER environment variable. For more information, see the CLI reference page.

Once the experiment is started, you will see a notification:

Preparing files (.../mnist_pytorch) to send to master... 2.5KB and 4 files
Created experiment xxx

Evaluating the Model

Model evaluation is done automatically for you by Determined. To access information on both training and validation performance, simply go to the WebUI by entering the address of the Determined master in your web browser.

Once you are on the Determined landing page, you can find your experiment using the experiment’s ID (xxx in the example above) or description.