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 31, 2023 | 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.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
import torchvision.transforms as transforms

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


transform = transforms.Compose(
    [transforms.ToTensor(),
    transforms.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('Training set has {} instances'.format(len(training_set)))
print('Validation set has {} instances'.format(len(validation_set)))

  0%|          | 0.00/26.4M [00:00<?, ?B/s]
  0%|          | 65.5k/26.4M [00:00<01:13, 361kB/s]
  1%|          | 229k/26.4M [00:00<00:38, 677kB/s]
  3%|▎         | 918k/26.4M [00:00<00:12, 2.09MB/s]
 14%|█▍        | 3.67M/26.4M [00:00<00:03, 7.22MB/s]
 38%|███▊      | 9.93M/26.4M [00:00<00:00, 17.0MB/s]
 56%|█████▌    | 14.8M/26.4M [00:01<00:00, 24.0MB/s]
 73%|███████▎  | 19.2M/26.4M [00:01<00:00, 24.4MB/s]
 89%|████████▉ | 23.5M/26.4M [00:01<00:00, 28.6MB/s]
100%|██████████| 26.4M/26.4M [00:01<00:00, 19.2MB/s]

  0%|          | 0.00/29.5k [00:00<?, ?B/s]
100%|██████████| 29.5k/29.5k [00:00<00:00, 326kB/s]

  0%|          | 0.00/4.42M [00:00<?, ?B/s]
  1%|▏         | 65.5k/4.42M [00:00<00:12, 362kB/s]
  5%|▌         | 229k/4.42M [00:00<00:06, 682kB/s]
 19%|█▉        | 852k/4.42M [00:00<00:01, 1.95MB/s]
 79%|███████▊  | 3.47M/4.42M [00:00<00:00, 6.89MB/s]
100%|██████████| 4.42M/4.42M [00:00<00:00, 6.10MB/s]

  0%|          | 0.00/5.15k [00:00<?, ?B/s]
100%|██████████| 5.15k/5.15k [00:00<00:00, 49.1MB/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)))

Ankle Boot  Coat  Shirt  Dress

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(GarmentClassifier, self).__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('Total loss for this batch: {}'.format(loss.item()))

tensor([[0.5312, 0.9144, 0.3713, 0.2023, 0.1512, 0.0514, 0.3545, 0.2707, 0.7593,
         0.1180],
        [0.4891, 0.1697, 0.5060, 0.3945, 0.7627, 0.9200, 0.2309, 0.0708, 0.7425,
         0.5815],
        [0.5070, 0.2311, 0.8580, 0.6680, 0.9576, 0.4998, 0.8674, 0.9511, 0.0168,
         0.8325],
        [0.0212, 0.8904, 0.8571, 0.4029, 0.0714, 0.7601, 0.0895, 0.7758, 0.2345,
         0.5122]])
tensor([1, 5, 3, 7])
Total loss for this batch: 2.0149483680725098

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('  batch {} loss: {}'.format(i + 1, 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('runs/fashion_trainer_{}'.format(timestamp))
epoch_number = 0

EPOCHS = 5

best_vloss = 1_000_000.

for epoch in range(EPOCHS):
    print('EPOCH {}:'.format(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('LOSS train {} valid {}'.format(avg_loss, 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 = 'model_{}_{}'.format(timestamp, epoch_number)
        torch.save(model.state_dict(), model_path)

    epoch_number += 1

EPOCH 1:
  batch 1000 loss: 1.911613073259592
  batch 2000 loss: 0.8206162689644844
  batch 3000 loss: 0.7102046648731921
  batch 4000 loss: 0.6288113839561119
  batch 5000 loss: 0.60175653554569
  batch 6000 loss: 0.5311959320821333
  batch 7000 loss: 0.5092687486642972
  batch 8000 loss: 0.49987754466210027
  batch 9000 loss: 0.4688529164947104
  batch 10000 loss: 0.46122014291386587
  batch 11000 loss: 0.4682306807874702
  batch 12000 loss: 0.46635702697257514
  batch 13000 loss: 0.4220156363591086
  batch 14000 loss: 0.41290630070699263
  batch 15000 loss: 0.40346511359675785
LOSS train 0.40346511359675785 valid 0.4482421278953552
EPOCH 2:
  batch 1000 loss: 0.3962256389490794
  batch 2000 loss: 0.41334549222965145
  batch 3000 loss: 0.40296541169908595
  batch 4000 loss: 0.3892910019377014
  batch 5000 loss: 0.37593005617469316
  batch 6000 loss: 0.3651836504072999
  batch 7000 loss: 0.3553595610834163
  batch 8000 loss: 0.33688234518578974
  batch 9000 loss: 0.3803561602494883
  batch 10000 loss: 0.37356412972530234
  batch 11000 loss: 0.36027771969515016
  batch 12000 loss: 0.35031046262569726
  batch 13000 loss: 0.3651484847535903
  batch 14000 loss: 0.37761624862375903
  batch 15000 loss: 0.3648304417096806
LOSS train 0.3648304417096806 valid 0.3588311970233917
EPOCH 3:
  batch 1000 loss: 0.3529048419316532
  batch 2000 loss: 0.31862275824469544
  batch 3000 loss: 0.314526914866743
  batch 4000 loss: 0.31931589686966616
  batch 5000 loss: 0.36001923132693625
  batch 6000 loss: 0.3253918188666576
  batch 7000 loss: 0.34011145413856136
  batch 8000 loss: 0.31314942210634034
  batch 9000 loss: 0.3279261530553631
  batch 10000 loss: 0.3613286251967802
  batch 11000 loss: 0.32283151171560165
  batch 12000 loss: 0.32569421616842736
  batch 13000 loss: 0.3297199288518459
  batch 14000 loss: 0.32277034278901917
  batch 15000 loss: 0.3152707609322315
LOSS train 0.3152707609322315 valid 0.33958348631858826
EPOCH 4:
  batch 1000 loss: 0.30576178838692614
  batch 2000 loss: 0.3179075249790185
  batch 3000 loss: 0.3028234139950655
  batch 4000 loss: 0.2912576552517785
  batch 5000 loss: 0.29363624689599965
  batch 6000 loss: 0.31323994945414596
  batch 7000 loss: 0.30978503643177646
  batch 8000 loss: 0.2899228875869558
  batch 9000 loss: 0.2864801618416605
  batch 10000 loss: 0.30417800577753223
  batch 11000 loss: 0.28836172214374345
  batch 12000 loss: 0.30941605619694745
  batch 13000 loss: 0.3221794830676954
  batch 14000 loss: 0.3253727466590608
  batch 15000 loss: 0.2932491722118284
LOSS train 0.2932491722118284 valid 0.31831732392311096
EPOCH 5:
  batch 1000 loss: 0.28041908680988625
  batch 2000 loss: 0.29547160665330646
  batch 3000 loss: 0.28921972356474723
  batch 4000 loss: 0.2885129695975884
  batch 5000 loss: 0.2812042115759323
  batch 6000 loss: 0.28170365359463906
  batch 7000 loss: 0.272492921995763
  batch 8000 loss: 0.29509544632599866
  batch 9000 loss: 0.27237311259275887
  batch 10000 loss: 0.29381787677724786
  batch 11000 loss: 0.2902567847310038
  batch 12000 loss: 0.26909997308760103
  batch 13000 loss: 0.2878994586916815
  batch 14000 loss: 0.2870736055240195
  batch 15000 loss: 0.29489504894919444
LOSS train 0.29489504894919444 valid 0.3201742470264435

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