DataParallel basics in PyTorch

Introduction

DataParallel helps your program use multiple GPUs to train a model faster by splitting the work.

You want to train a neural network faster using more than one GPU.

Your model fits in one GPU but you want to use extra GPUs to speed up training.

You have a desktop or server with multiple GPUs and want to use them easily.

You want to keep your code simple while using multiple GPUs.

Syntax

PyTorch

model = torch.nn.DataParallel(model)
output = model(input)

Wrap your model with DataParallel before training.

Input data is automatically split across GPUs.

Examples

Wrap a simple linear model to use DataParallel.

PyTorch

import torch
import torch.nn as nn

model = nn.Linear(10, 5)
model = torch.nn.DataParallel(model)

Pass a batch of 16 samples to the parallel model.

PyTorch

output = model(torch.randn(16, 10))

Sample Model

This code creates a simple model, wraps it with DataParallel if multiple GPUs are available, and runs one training step with dummy data.

PyTorch

import torch
import torch.nn as nn
import torch.optim as optim

# Simple model
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(10, 2)
    def forward(self, x):
        return self.linear(x)

# Check if multiple GPUs are available
if torch.cuda.device_count() > 1:
    print(f"Using {torch.cuda.device_count()} GPUs")
else:
    print("Using single GPU or CPU")

# Create model and move to GPUs
model = SimpleModel()
if torch.cuda.is_available():
    model = model.cuda()
    if torch.cuda.device_count() > 1:
        model = nn.DataParallel(model)

# Dummy data
inputs = torch.randn(32, 10)
labels = torch.randint(0, 2, (32,))
if torch.cuda.is_available():
    inputs = inputs.cuda()
    labels = labels.cuda()

# Loss and optimizer
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(model.parameters(), lr=0.01)

# Forward pass
outputs = model(inputs)
loss = criterion(outputs, labels)

# Backward and optimize
optimizer.zero_grad()
loss.backward()
optimizer.step()

print(f"Loss: {loss.item():.4f}")

Important Notes

DataParallel splits input batches automatically across GPUs.

Model outputs are combined back on the main GPU.

DataParallel is easy but not the fastest multi-GPU method; consider DistributedDataParallel for large projects.

Summary

DataParallel lets you use multiple GPUs easily by wrapping your model.

It splits input data and combines outputs automatically.

Good for beginners to speed up training without complex code changes.

Practice

1. What is the main purpose of torch.nn.DataParallel in PyTorch?

easy

A. To convert a model to run only on CPU

B. To run a model on multiple GPUs by splitting input data automatically

C. To reduce the size of the model parameters

D. To change the model architecture dynamically during training

Solution

1. Step 1: Understand DataParallel's role

   DataParallel helps use multiple GPUs by splitting input data and running model copies in parallel.

2. Step 2: Compare options

   Only To run a model on multiple GPUs by splitting input data automatically describes this behavior correctly; others describe unrelated tasks.

3. Final Answer:

   To run a model on multiple GPUs by splitting input data automatically -> Option B

4. Quick Check:

   DataParallel = multi-GPU input splitting [OK]

Hint: DataParallel splits input across GPUs automatically [OK]

Common Mistakes:

• Thinking DataParallel changes model architecture
• Confusing DataParallel with CPU-only mode
• Believing it reduces model size

2. Which of the following is the correct way to wrap a PyTorch model model with DataParallel?

easy

A. model = torch.DataParallel(model)

B. model = torch.nn.parallel.DataParallel(model)

C. model = torch.nn.DataParallel(model)

D. model = torch.parallel.DataParallel(model)

Solution

1. Step 1: Recall correct import path

   DataParallel is in torch.nn, so correct usage is torch.nn.DataParallel.

2. Step 2: Check options

   model = torch.nn.DataParallel(model) matches the correct usage; others are incorrect or incomplete.

3. Final Answer:

   model = torch.nn.DataParallel(model) -> Option C

4. Quick Check:

   Correct import path = torch.nn.DataParallel [OK]

Hint: DataParallel is in torch.nn module [OK]

Common Mistakes:

• Using torch.DataParallel instead of torch.nn.DataParallel
• Confusing torch.nn.parallel with torch.nn
• Confusing with other torch submodules

3. Given the code below, what will be the output shape of output if input shape is (8, 3, 32, 32) and 2 GPUs are used?

import torch
import torch.nn as nn

model = nn.Linear(32*32*3, 10)
model = nn.parallel.DataParallel(model)
input = torch.randn(8, 3, 32, 32)
input = input.view(8, -1)
output = model(input)
print(output.shape)

medium

A. torch.Size([8, 10])

B. torch.Size([4, 10])

C. torch.Size([16, 10])

D. RuntimeError due to shape mismatch

Solution

1. Step 1: Understand input and model

   Input is reshaped to (8, 3072) and model outputs 10 features per input.

2. Step 2: DataParallel splits batch across 2 GPUs

   Batch of 8 splits into 2 batches of 4, processed in parallel, outputs combined back to (8, 10).

3. Final Answer:

   torch.Size([8, 10]) -> Option A

4. Quick Check:

   Output shape = batch size x output features = (8, 10) [OK]

Hint: Output batch size stays same after DataParallel [OK]

Common Mistakes:

• Thinking output batch size halves with GPUs
• Confusing input reshape with output shape
• Expecting runtime error due to DataParallel

4. What is the likely cause of this error when using DataParallel?

RuntimeError: Expected tensor for argument #1 'input' to have scalar type Float but got Double

Code snippet:

model = torch.nn.parallel.DataParallel(model)
input = torch.randn(8, 3, 32, 32, dtype=torch.double)
output = model(input)

medium

A. Input batch size is too large for GPUs

B. DataParallel requires input to be on CPU only

C. Model is not wrapped correctly with DataParallel

D. Input tensor dtype does not match model parameters dtype

Solution

1. Step 1: Analyze error message

   Error says input tensor is Double but model expects Float tensors.

2. Step 2: Check input and model dtypes

   Model parameters default to Float; input is Double, causing mismatch and error.

3. Final Answer:

   Input tensor dtype does not match model parameters dtype -> Option D

4. Quick Check:

   Dtype mismatch causes runtime error [OK]

Hint: Ensure input and model dtypes match before DataParallel [OK]

Common Mistakes:

• Assuming DataParallel only works on CPU
• Not checking tensor data types
• Thinking batch size causes dtype errors

5. You want to speed up training on 4 GPUs using DataParallel. Your model has a batch normalization layer. What should you be careful about to avoid incorrect training behavior?

hard

A. BatchNorm layers behave differently in DataParallel; use SyncBatchNorm or move to DistributedDataParallel

B. DataParallel automatically fixes BatchNorm issues, no action needed

C. Remove BatchNorm layers before using DataParallel

D. Increase batch size to avoid BatchNorm problems

Solution

1. Step 1: Understand BatchNorm behavior in DataParallel

   BatchNorm computes stats per device, which can cause inconsistent normalization across GPUs.

2. Step 2: Recommended solutions

   Use SyncBatchNorm to sync stats or switch to DistributedDataParallel for correct BatchNorm behavior.

3. Final Answer:

   BatchNorm layers behave differently in DataParallel; use SyncBatchNorm or move to DistributedDataParallel -> Option A

4. Quick Check:

   BatchNorm needs syncing or DDP for multi-GPU [OK]

Hint: BatchNorm needs syncing or DDP on multiple GPUs [OK]

Common Mistakes:

• Assuming DataParallel handles BatchNorm correctly by default
• Removing BatchNorm layers unnecessarily
• Only increasing batch size without syncing stats