beginner/introyt/trainingyt

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

Introduction || Tensors || Autograd || Building Models || TensorBoard Support || Training Models || Model Understanding

Training with PyTorch

Created On: Nov 30, 2021 | Last Updated: May 06, 2026 | Last Verified: Nov 05, 2024

Follow along with the video below or on youtube.

Introduction

In past videos, we’ve discussed and demonstrated:

• Building models with the neural network layers and functions of the torch.nn module

• The mechanics of automated gradient computation, which is central to gradient-based model training

• Using TensorBoard to visualize training progress and other activities

In this video, we’ll be adding some new tools to your inventory:

• We’ll get familiar with the dataset and dataloader abstractions, and how they ease the process of feeding data to your model during a training loop

• We’ll discuss specific loss functions and when to use them

• We’ll look at PyTorch optimizers, which implement algorithms to adjust model weights based on the outcome of a loss function

Finally, we’ll pull all of these together and see a full PyTorch training loop in action.

Dataset and DataLoader

The Dataset and DataLoader classes encapsulate the process of pulling your data from storage and exposing it to your training loop in batches.

The Dataset is responsible for accessing and processing single instances of data.

The DataLoader pulls instances of data from the Dataset (either automatically or with a sampler that you define), collects them in batches, and returns them for consumption by your training loop. The DataLoader works with all kinds of datasets, regardless of the type of data they contain.

For this tutorial, we’ll be using the Fashion-MNIST dataset provided by TorchVision. We use torchvision.transforms.v2.Normalize() to zero-center and normalize the distribution of the image tile content, and download both training and validation data splits.

import torch
import torchvision
from torchvision.transforms import v2

# PyTorch TensorBoard support
from torch.utils.tensorboard import SummaryWriter
from datetime import datetime


transform = v2.Compose([
    v2.ToImage(),
    v2.ToDtype(torch.float32, scale=True),
    v2.Normalize((0.5,), (0.5,))
])

# Create datasets for training & validation, download if necessary
training_set = torchvision.datasets.FashionMNIST('./data', train=True, transform=transform, download=True)
validation_set = torchvision.datasets.FashionMNIST('./data', train=False, transform=transform, download=True)

# Create data loaders for our datasets; shuffle for training, not for validation
training_loader = torch.utils.data.DataLoader(training_set, batch_size=4, shuffle=True)
validation_loader = torch.utils.data.DataLoader(validation_set, batch_size=4, shuffle=False)

# Class labels
classes = ('T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat',
        'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle Boot')

# Report split sizes
print(f'Training set has {len(training_set)} instances')
print(f'Validation set has {len(validation_set)} instances')

  0%|          | 0.00/26.4M [00:00<?, ?B/s]
  0%|          | 65.5k/26.4M [00:00<01:09, 380kB/s]
  1%|          | 229k/26.4M [00:00<00:36, 712kB/s]
  3%|▎         | 918k/26.4M [00:00<00:11, 2.20MB/s]
 14%|█▎        | 3.60M/26.4M [00:00<00:02, 8.27MB/s]
 26%|██▌       | 6.78M/26.4M [00:00<00:01, 13.4MB/s]
 48%|████▊     | 12.7M/26.4M [00:00<00:00, 23.4MB/s]
 61%|██████    | 16.1M/26.4M [00:01<00:00, 24.3MB/s]
 83%|████████▎ | 21.9M/26.4M [00:01<00:00, 30.4MB/s]
 96%|█████████▌| 25.4M/26.4M [00:01<00:00, 29.3MB/s]
100%|██████████| 26.4M/26.4M [00:01<00:00, 20.2MB/s]

  0%|          | 0.00/29.5k [00:00<?, ?B/s]
100%|██████████| 29.5k/29.5k [00:00<00:00, 330kB/s]

  0%|          | 0.00/4.42M [00:00<?, ?B/s]
  1%|▏         | 65.5k/4.42M [00:00<00:11, 371kB/s]
  5%|▌         | 229k/4.42M [00:00<00:05, 700kB/s]
 21%|██        | 918k/4.42M [00:00<00:01, 2.16MB/s]
 83%|████████▎ | 3.67M/4.42M [00:00<00:00, 7.47MB/s]
100%|██████████| 4.42M/4.42M [00:00<00:00, 6.26MB/s]

  0%|          | 0.00/5.15k [00:00<?, ?B/s]
100%|██████████| 5.15k/5.15k [00:00<00:00, 55.4MB/s]
Training set has 60000 instances
Validation set has 10000 instances

As always, let’s visualize the data as a sanity check:

import matplotlib.pyplot as plt
import numpy as np

# Helper function for inline image display
def matplotlib_imshow(img, one_channel=False):
    if one_channel:
        img = img.mean(dim=0)
    img = img / 2 + 0.5     # unnormalize
    npimg = img.numpy()
    if one_channel:
        plt.imshow(npimg, cmap="Greys")
    else:
        plt.imshow(np.transpose(npimg, (1, 2, 0)))

dataiter = iter(training_loader)
images, labels = next(dataiter)

# Create a grid from the images and show them
img_grid = torchvision.utils.make_grid(images)
matplotlib_imshow(img_grid, one_channel=True)
print('  '.join(classes[labels[j]] for j in range(4)))

Pullover  Coat  Dress  Pullover

The Model

The model we’ll use in this example is a variant of LeNet-5 - it should be familiar if you’ve watched the previous videos in this series.

import torch.nn as nn
import torch.nn.functional as F

# PyTorch models inherit from torch.nn.Module
class GarmentClassifier(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 6, 5)
        self.pool = nn.MaxPool2d(2, 2)
        self.conv2 = nn.Conv2d(6, 16, 5)
        self.fc1 = nn.Linear(16 * 4 * 4, 120)
        self.fc2 = nn.Linear(120, 84)
        self.fc3 = nn.Linear(84, 10)

    def forward(self, x):
        x = self.pool(F.relu(self.conv1(x)))
        x = self.pool(F.relu(self.conv2(x)))
        x = x.view(-1, 16 * 4 * 4)
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x


model = GarmentClassifier()

Loss Function

For this example, we’ll be using a cross-entropy loss. For demonstration purposes, we’ll create batches of dummy output and label values, run them through the loss function, and examine the result.

loss_fn = torch.nn.CrossEntropyLoss()

# NB: Loss functions expect data in batches, so we're creating batches of 4
# Represents the model's confidence in each of the 10 classes for a given input
dummy_outputs = torch.rand(4, 10)
# Represents the correct class among the 10 being tested
dummy_labels = torch.tensor([1, 5, 3, 7])

print(dummy_outputs)
print(dummy_labels)

loss = loss_fn(dummy_outputs, dummy_labels)
print(f'Total loss for this batch: {loss.item()}')

tensor([[0.0382, 0.8669, 0.7359, 0.6873, 0.6889, 0.6340, 0.8957, 0.8771, 0.6930,
         0.9088],
        [0.6315, 0.5340, 0.1269, 0.5319, 0.1339, 0.7312, 0.2028, 0.7645, 0.2659,
         0.0300],
        [0.0116, 0.2622, 0.5236, 0.9573, 0.0947, 0.7915, 0.2107, 0.7610, 0.3983,
         0.2272],
        [0.3765, 0.9635, 0.7743, 0.3082, 0.4414, 0.7872, 0.1299, 0.1826, 0.4251,
         0.8568]])
tensor([1, 5, 3, 7])
Total loss for this batch: 2.1662769317626953

Optimizer

For this example, we’ll be using simple stochastic gradient descent with momentum.

It can be instructive to try some variations on this optimization scheme:

• Learning rate determines the size of the steps the optimizer takes. What does a different learning rate do to the your training results, in terms of accuracy and convergence time?

• Momentum nudges the optimizer in the direction of strongest gradient over multiple steps. What does changing this value do to your results?

• Try some different optimization algorithms, such as averaged SGD, Adagrad, or Adam. How do your results differ?

# Optimizers specified in the torch.optim package
optimizer = torch.optim.SGD(model.parameters(), lr=0.001, momentum=0.9)

The Training Loop

Below, we have a function that performs one training epoch. It enumerates data from the DataLoader, and on each pass of the loop does the following:

• Gets a batch of training data from the DataLoader

• Zeros the optimizer’s gradients

• Performs an inference - that is, gets predictions from the model for an input batch

• Calculates the loss for that set of predictions vs. the labels on the dataset

• Calculates the backward gradients over the learning weights

• Tells the optimizer to perform one learning step - that is, adjust the model’s learning weights based on the observed gradients for this batch, according to the optimization algorithm we chose

• It reports on the loss for every 1000 batches.

• Finally, it reports the average per-batch loss for the last 1000 batches, for comparison with a validation run

def train_one_epoch(epoch_index, tb_writer):
    running_loss = 0.
    last_loss = 0.

    # Here, we use enumerate(training_loader) instead of
    # iter(training_loader) so that we can track the batch
    # index and do some intra-epoch reporting
    for i, data in enumerate(training_loader):
        # Every data instance is an input + label pair
        inputs, labels = data

        # Zero your gradients for every batch!
        optimizer.zero_grad()

        # Make predictions for this batch
        outputs = model(inputs)

        # Compute the loss and its gradients
        loss = loss_fn(outputs, labels)
        loss.backward()

        # Adjust learning weights
        optimizer.step()

        # Gather data and report
        running_loss += loss.item()
        if i % 1000 == 999:
            last_loss = running_loss / 1000 # loss per batch
            print(f'  batch {i + 1} loss: {last_loss}')
            tb_x = epoch_index * len(training_loader) + i + 1
            tb_writer.add_scalar('Loss/train', last_loss, tb_x)
            running_loss = 0.

    return last_loss

Per-Epoch Activity

There are a couple of things we’ll want to do once per epoch:

• Perform validation by checking our relative loss on a set of data that was not used for training, and report this

• Save a copy of the model

Here, we’ll do our reporting in TensorBoard. This will require going to the command line to start TensorBoard, and opening it in another browser tab.

# Initializing in a separate cell so we can easily add more epochs to the same run
timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
writer = SummaryWriter(f'runs/fashion_trainer_{timestamp}')
epoch_number = 0

EPOCHS = 5

best_vloss = 1_000_000.

for epoch in range(EPOCHS):
    print(f'EPOCH {epoch_number + 1}:')

    # Make sure gradient tracking is on, and do a pass over the data
    model.train(True)
    avg_loss = train_one_epoch(epoch_number, writer)


    running_vloss = 0.0
    # Set the model to evaluation mode, disabling dropout and using population
    # statistics for batch normalization.
    model.eval()

    # Disable gradient computation and reduce memory consumption.
    with torch.no_grad():
        for i, vdata in enumerate(validation_loader):
            vinputs, vlabels = vdata
            voutputs = model(vinputs)
            vloss = loss_fn(voutputs, vlabels)
            running_vloss += vloss

    avg_vloss = running_vloss / (i + 1)
    print(f'LOSS train {avg_loss} valid {avg_vloss}')

    # Log the running loss averaged per batch
    # for both training and validation
    writer.add_scalars('Training vs. Validation Loss',
                    { 'Training' : avg_loss, 'Validation' : avg_vloss },
                    epoch_number + 1)
    writer.flush()

    # Track best performance, and save the model's state
    if avg_vloss < best_vloss:
        best_vloss = avg_vloss
        model_path = f'model_{timestamp}_{epoch_number}'
        torch.save(model.state_dict(), model_path)

    epoch_number += 1

EPOCH 1:
  batch 1000 loss: 2.049647333547473
  batch 2000 loss: 0.9055713153481484
  batch 3000 loss: 0.7383996275421232
  batch 4000 loss: 0.6548138484992087
  batch 5000 loss: 0.591857960563153
  batch 6000 loss: 0.5675007280143909
  batch 7000 loss: 0.54833523921913
  batch 8000 loss: 0.5285603157215519
  batch 9000 loss: 0.5111474491535045
  batch 10000 loss: 0.47651419608923606
  batch 11000 loss: 0.4836354687914718
  batch 12000 loss: 0.4574846286987886
  batch 13000 loss: 0.4338111500995001
  batch 14000 loss: 0.4434925923334668
  batch 15000 loss: 0.4387456377467606
LOSS train 0.4387456377467606 valid 0.4373689591884613
EPOCH 2:
  batch 1000 loss: 0.42765470422612273
  batch 2000 loss: 0.39657323571524467
  batch 3000 loss: 0.3999155504635128
  batch 4000 loss: 0.38150215953204314
  batch 5000 loss: 0.38767915842781075
  batch 6000 loss: 0.37517543748847676
  batch 7000 loss: 0.3778295161024435
  batch 8000 loss: 0.37112019500200405
  batch 9000 loss: 0.3880192909524194
  batch 10000 loss: 0.34236860253626944
  batch 11000 loss: 0.3589225430016522
  batch 12000 loss: 0.3576424698217088
  batch 13000 loss: 0.33499407355097355
  batch 14000 loss: 0.3536516610257095
  batch 15000 loss: 0.3257400691052317
LOSS train 0.3257400691052317 valid 0.3763027489185333
EPOCH 3:
  batch 1000 loss: 0.3378695148827028
  batch 2000 loss: 0.32500762943126027
  batch 3000 loss: 0.3205619038755103
  batch 4000 loss: 0.34099642446293726
  batch 5000 loss: 0.3248557770683255
  batch 6000 loss: 0.3287246027403744
  batch 7000 loss: 0.3013888432139356
  batch 8000 loss: 0.312386061328958
  batch 9000 loss: 0.33087204418840704
  batch 10000 loss: 0.3231138217273692
  batch 11000 loss: 0.3299627491485226
  batch 12000 loss: 0.3101591057227779
  batch 13000 loss: 0.3313205327987089
  batch 14000 loss: 0.31772167054853345
  batch 15000 loss: 0.3110737975396041
LOSS train 0.3110737975396041 valid 0.3413317799568176
EPOCH 4:
  batch 1000 loss: 0.29113358351755597
  batch 2000 loss: 0.2982887633475475
  batch 3000 loss: 0.2999963893459426
  batch 4000 loss: 0.29396984903886914
  batch 5000 loss: 0.32292703295339015
  batch 6000 loss: 0.2855030224823713
  batch 7000 loss: 0.3092117739984533
  batch 8000 loss: 0.2866465583481622
  batch 9000 loss: 0.29277199014558575
  batch 10000 loss: 0.28194521764779346
  batch 11000 loss: 0.29269337059125244
  batch 12000 loss: 0.2957005590953049
  batch 13000 loss: 0.3029561814356166
  batch 14000 loss: 0.2963960933635899
  batch 15000 loss: 0.30151225167872325
LOSS train 0.30151225167872325 valid 0.3140939772129059
EPOCH 5:
  batch 1000 loss: 0.2755536031205702
  batch 2000 loss: 0.2758934389857168
  batch 3000 loss: 0.27942894034357596
  batch 4000 loss: 0.28048185547199683
  batch 5000 loss: 0.2721731190502369
  batch 6000 loss: 0.2797486772877601
  batch 7000 loss: 0.2925731274943464
  batch 8000 loss: 0.30628264715825027
  batch 9000 loss: 0.26380198201824534
  batch 10000 loss: 0.2838711646141401
  batch 11000 loss: 0.2715263127603685
  batch 12000 loss: 0.2880003579687909
  batch 13000 loss: 0.27976815830593116
  batch 14000 loss: 0.2613652834414752
  batch 15000 loss: 0.25735641520857006
LOSS train 0.25735641520857006 valid 0.3274642825126648

To load a saved version of the model:

saved_model = GarmentClassifier()
saved_model.load_state_dict(torch.load(PATH))

Once you’ve loaded the model, it’s ready for whatever you need it for - more training, inference, or analysis.

Note that if your model has constructor parameters that affect model structure, you’ll need to provide them and configure the model identically to the state in which it was saved.

Other Resources

• Docs on the data utilities, including Dataset and DataLoader, at pytorch.org

• A note on the use of pinned memory for GPU training

• Documentation on the datasets available in TorchVision, TorchText, and TorchAudio

• Documentation on the loss functions available in PyTorch

• Documentation on the torch.optim package, which includes optimizers and related tools, such as learning rate scheduling

• A detailed tutorial on saving and loading models

• The Tutorials section of pytorch.org contains tutorials on a broad variety of training tasks, including classification in different domains, generative adversarial networks, reinforcement learning, and more