ONNX export in PyTorch

Introduction

ONNX export lets you save your PyTorch model in a format that many other tools and platforms can use. This helps share and run your model outside PyTorch easily.

You want to run your PyTorch model on a different system that does not support PyTorch.

You want to optimize your model for faster inference using ONNX runtime or other accelerators.

You want to deploy your model to mobile or web platforms that support ONNX.

You want to convert your model to another framework like TensorFlow or Caffe2.

You want to share your model with others who use different machine learning tools.

Syntax

PyTorch

torch.onnx.export(model, args, f, export_params=True, opset_version=None, do_constant_folding=True, input_names=None, output_names=None, dynamic_axes=None)

model: Your trained PyTorch model.

args: Example input tensor(s) to trace the model.

f: File path to save the ONNX model.

Examples

Export model with default settings using a dummy input tensor.

PyTorch

torch.onnx.export(model, dummy_input, "model.onnx")

Specify names for input and output nodes for clarity.

PyTorch

torch.onnx.export(model, dummy_input, "model.onnx", input_names=["input"], output_names=["output"])

Export with opset version 11 and dynamic batch size support.

PyTorch

torch.onnx.export(model, dummy_input, "model.onnx", opset_version=11, dynamic_axes={"input": {0: "batch_size"}, "output": {0: "batch_size"}})

Sample Model

This code defines a small linear model, creates a dummy input, and exports the model to ONNX format with named inputs and outputs. It also supports dynamic batch size.

PyTorch

import torch
import torch.nn as nn

# Define a simple model
class SimpleModel(nn.Module):
    def __init__(self):
        super(SimpleModel, self).__init__()
        self.linear = nn.Linear(3, 2)
    def forward(self, x):
        return self.linear(x)

model = SimpleModel()
model.eval()  # Set to evaluation mode

dummy_input = torch.randn(1, 3)  # Example input

# Export the model to ONNX format
torch.onnx.export(
    model,
    dummy_input,
    "simple_model.onnx",
    input_names=["input"],
    output_names=["output"],
    opset_version=11,
    dynamic_axes={"input": {0: "batch_size"}, "output": {0: "batch_size"}}
)

print("ONNX model exported successfully to simple_model.onnx")

Important Notes

Make sure your model is in evaluation mode (model.eval()) before exporting to avoid training-only behaviors like dropout.

Use a dummy input tensor with the correct shape to trace the model correctly.

Choosing the right opset_version ensures compatibility with the ONNX runtime or other tools you plan to use.

Summary

ONNX export saves PyTorch models in a universal format for sharing and deployment.

Use dummy inputs and set model.eval() before exporting.

You can customize input/output names and support dynamic shapes during export.

Practice

1. What is the main purpose of exporting a PyTorch model to ONNX format?

easy

A. To save the model in a universal format for sharing and deployment

B. To train the model faster on GPUs

C. To convert the model into a TensorFlow model automatically

D. To visualize the model architecture in PyTorch

Solution

1. Step 1: Understand ONNX export purpose

   ONNX is designed to save models in a format that can be used across different frameworks and platforms.

2. Step 2: Compare options

   Only To save the model in a universal format for sharing and deployment correctly describes the universal sharing and deployment purpose of ONNX export.

3. Final Answer:

   To save the model in a universal format for sharing and deployment -> Option A

4. Quick Check:

   ONNX export = universal format [OK]

Hint: ONNX = share and deploy models universally [OK]

Common Mistakes:

• Confusing ONNX export with training speedup
• Thinking ONNX converts models to TensorFlow automatically
• Assuming ONNX export is for visualization only

2. Which of the following is the correct way to prepare a PyTorch model for ONNX export?

easy

A. Call model.train() before export

B. Export without setting any input

C. Use a dummy input tensor matching the model input shape

D. Use model.eval() after export

Solution

1. Step 1: Identify preparation steps for ONNX export

   Model should be in evaluation mode and a dummy input tensor matching input shape is needed for tracing.

2. Step 2: Evaluate options

   Use a dummy input tensor matching the model input shape correctly states the use of a dummy input tensor. Call model.train() before export is wrong because model.train() is for training mode. Export without setting any input misses input, and Use model.eval() after export is incorrect order.

3. Final Answer:

   Use a dummy input tensor matching the model input shape -> Option C

4. Quick Check:

   Dummy input needed = Use a dummy input tensor matching the model input shape [OK]

Hint: Always use dummy input tensor before export [OK]

Common Mistakes:

• Forgetting to set model.eval() before export
• Not providing dummy input tensor
• Calling model.train() instead of eval()

3. Given the code below, what will be the output of print(onnx_model.graph.input[0].name) after export?

import torch
import torch.nn as nn

class SimpleModel(nn.Module):
    def forward(self, x):
        return x * 2

model = SimpleModel()
model.eval()
dummy_input = torch.randn(1, 3)
torch.onnx.export(model, dummy_input, "model.onnx", input_names=["input_tensor"])
import onnx
onnx_model = onnx.load("model.onnx")
print(onnx_model.graph.input[0].name)

medium

A. x

B. input_tensor

C. input0

D. data

Solution

1. Step 1: Check input_names parameter in export

   The export call sets input_names=["input_tensor"], so the input name in ONNX graph should be "input_tensor".

2. Step 2: Confirm printed input name

   Loading the ONNX model and printing the first input name will output "input_tensor" as set.

3. Final Answer:

   input_tensor -> Option B

4. Quick Check:

   input_names param = input_tensor [OK]

Hint: input_names param sets ONNX input name [OK]

Common Mistakes:

• Assuming default input name like 'input0'
• Confusing PyTorch variable name with ONNX input name
• Not setting input_names and expecting custom name

4. You try to export a PyTorch model to ONNX but get an error: RuntimeError: Input type (torch.cuda.FloatTensor) and weight type (torch.FloatTensor) should be the same. What is the best fix?

medium

A. Export without specifying input_names

B. Use model.train() instead of model.eval()

C. Remove dummy input tensor

D. Move both the model and the dummy input to CPU before export

Solution

1. Step 1: Understand the error cause

   The error means model weights are on CPU but input tensor is on GPU (cuda), causing type mismatch.

2. Step 2: Fix by aligning device

   Moving both the model and the dummy input to CPU ensures both are on the same device (CPU), fixing the mismatch.

3. Final Answer:

   Move both the model and the dummy input to CPU before export -> Option D

4. Quick Check:

   Device mismatch fix = move model and input to CPU [OK]

Hint: Ensure model and input are on same device before export [OK]

Common Mistakes:

• Ignoring device mismatch and exporting anyway
• Switching to train mode instead of fixing device
• Removing dummy input tensor causing other errors

5. You want to export a PyTorch model to ONNX with dynamic batch size support. Which argument should you add to torch.onnx.export to enable this?

hard

A. dynamic_axes={'input': {0: 'batch_size'}, 'output': {0: 'batch_size'}}

B. enable_dynamic_batch=True

C. set_dynamic=True

D. dynamic_batch=True

Solution

1. Step 1: Identify how to specify dynamic axes in ONNX export

   PyTorch ONNX export uses the dynamic_axes argument to mark dimensions as dynamic, e.g., batch size dimension 0.

2. Step 2: Check options for dynamic batch size

   Only dynamic_axes={'input': {0: 'batch_size'}, 'output': {0: 'batch_size'}} correctly uses dynamic_axes with dictionary specifying batch dimension 0 for input and output.

3. Final Answer:

   dynamic_axes={'input': {0: 'batch_size'}, 'output': {0: 'batch_size'}} -> Option A

4. Quick Check:

   Dynamic batch size = dynamic_axes param [OK]

Hint: Use dynamic_axes dict to set dynamic batch size [OK]

Common Mistakes:

• Using nonexistent parameters like enable_dynamic_batch
• Forgetting to specify dynamic axes for output
• Assuming batch size is dynamic by default