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<01:12, 364kB/s]
  1%|          | 229k/26.4M [00:00<00:38, 686kB/s]
  3%|▎         | 885k/26.4M [00:00<00:12, 2.03MB/s]
 14%|█▎        | 3.57M/26.4M [00:00<00:03, 7.11MB/s]
 36%|███▌      | 9.40M/26.4M [00:00<00:01, 16.2MB/s]
 58%|█████▊    | 15.2M/26.4M [00:01<00:00, 21.6MB/s]
 81%|████████  | 21.4M/26.4M [00:01<00:00, 25.7MB/s]
100%|██████████| 26.4M/26.4M [00:01<00:00, 19.4MB/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, 360kB/s]
  4%|▍         | 197k/4.42M [00:00<00:07, 573kB/s]
 19%|█▉        | 852k/4.42M [00:00<00:01, 1.95MB/s]
 76%|███████▋  | 3.38M/4.42M [00:00<00:00, 6.66MB/s]
100%|██████████| 4.42M/4.42M [00:00<00:00, 6.05MB/s]

  0%|          | 0.00/5.15k [00:00<?, ?B/s]
100%|██████████| 5.15k/5.15k [00:00<00:00, 51.5MB/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)))

Sneaker  Sandal  Sneaker  T-shirt/top

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([[0.7969, 0.7584, 0.7556, 0.2980, 0.8937, 0.6328, 0.8907, 0.3791, 0.5392,
         0.9928],
        [0.0435, 0.0696, 0.0679, 0.7475, 0.2197, 0.7399, 0.5288, 0.3662, 0.4188,
         0.1029],
        [0.4058, 0.2525, 0.1435, 0.2725, 0.0114, 0.3252, 0.7358, 0.5453, 0.1244,
         0.2116],
        [0.6904, 0.9058, 0.3586, 0.2889, 0.7449, 0.5637, 0.1873, 0.8514, 0.1016,
         0.9492]])
tensor([1, 5, 3, 7])
Total loss for this batch: 2.1500563621520996

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: 1.9389974682629108
  batch 2000 loss: 0.8669686885476112
  batch 3000 loss: 0.7235826873648912
  batch 4000 loss: 0.6343555997302756
  batch 5000 loss: 0.6038945620488375
  batch 6000 loss: 0.5301300394203281
  batch 7000 loss: 0.5329250979589997
  batch 8000 loss: 0.49135314543778075
  batch 9000 loss: 0.4941273945478024
  batch 10000 loss: 0.49375148038030603
  batch 11000 loss: 0.467604599519982
  batch 12000 loss: 0.44371729486389083
  batch 13000 loss: 0.4469605188059504
  batch 14000 loss: 0.4296329754444596
  batch 15000 loss: 0.424329058547446
LOSS train 0.424329058547446 valid 0.40960893034935
EPOCH 2:
  batch 1000 loss: 0.4086844542364124
  batch 2000 loss: 0.4004356552777172
  batch 3000 loss: 0.3822158178684476
  batch 4000 loss: 0.38350689092453105
  batch 5000 loss: 0.3982887527482235
  batch 6000 loss: 0.38413116014696425
  batch 7000 loss: 0.390030125937541
  batch 8000 loss: 0.4122948635958601
  batch 9000 loss: 0.39848179402842654
  batch 10000 loss: 0.3605497012230917
  batch 11000 loss: 0.3744867103410652
  batch 12000 loss: 0.3572737548819859
  batch 13000 loss: 0.354907838467916
  batch 14000 loss: 0.3465411294546502
  batch 15000 loss: 0.3465292985017004
LOSS train 0.3465292985017004 valid 0.38021528720855713
EPOCH 3:
  batch 1000 loss: 0.3512419121793355
  batch 2000 loss: 0.3322464874680154
  batch 3000 loss: 0.32956673602684167
  batch 4000 loss: 0.361940099674568
  batch 5000 loss: 0.3324062218151812
  batch 6000 loss: 0.3366893311944441
  batch 7000 loss: 0.3421411258783046
  batch 8000 loss: 0.32158286139927805
  batch 9000 loss: 0.33464859428604543
  batch 10000 loss: 0.3321187043938262
  batch 11000 loss: 0.3235943128764702
  batch 12000 loss: 0.3383171066781215
  batch 13000 loss: 0.33102326372133395
  batch 14000 loss: 0.3244007411996863
  batch 15000 loss: 0.3266687692314554
LOSS train 0.3266687692314554 valid 0.34519803524017334
EPOCH 4:
  batch 1000 loss: 0.3193929187379108
  batch 2000 loss: 0.30989905784811705
  batch 3000 loss: 0.31518700145318873
  batch 4000 loss: 0.320427497327546
  batch 5000 loss: 0.3143957124307453
  batch 6000 loss: 0.32729043563588855
  batch 7000 loss: 0.3032689992305386
  batch 8000 loss: 0.30959021030861184
  batch 9000 loss: 0.30493444365386674
  batch 10000 loss: 0.30126625370302645
  batch 11000 loss: 0.31425340176874306
  batch 12000 loss: 0.3048126271425572
  batch 13000 loss: 0.295037083029747
  batch 14000 loss: 0.29859171077071367
  batch 15000 loss: 0.30028764154974485
LOSS train 0.30028764154974485 valid 0.3249934911727905
EPOCH 5:
  batch 1000 loss: 0.28815486478729324
  batch 2000 loss: 0.29075736585423145
  batch 3000 loss: 0.2881369464831587
  batch 4000 loss: 0.2893028177918532
  batch 5000 loss: 0.309677734099987
  batch 6000 loss: 0.2918114792679771
  batch 7000 loss: 0.28364818626622673
  batch 8000 loss: 0.303782212409209
  batch 9000 loss: 0.27787542383548863
  batch 10000 loss: 0.2828599879331887
  batch 11000 loss: 0.29596336167014303
  batch 12000 loss: 0.3140515244587259
  batch 13000 loss: 0.28954496164216836
  batch 14000 loss: 0.2696223817095815
  batch 15000 loss: 0.28856453035336016
LOSS train 0.28856453035336016 valid 0.3206747770309448

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