Mobile deployment (PyTorch Mobile) - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to load a PyTorch model for mobile deployment.

PyTorch

import torch

model = torch.jit.load([1])
model.eval()

Drag options to blanks, or click blank then click option'

A"model.pt"

Bmodel.pt

Ctorch_model.pt

D'model.pth'

Attempts:

3 left

💡 Hint

Common Mistakes

Forgetting quotes around the file path string.

Using the wrong file extension like '.pth' instead of '.pt'.

The model file path must be a string with quotes, like "model.pt".

2fill in blank

medium

Complete the code to convert a PyTorch model to TorchScript for mobile deployment.

PyTorch

import torch

scripted_model = torch.jit.[1](model)
scripted_model.save("model.pt")

Drag options to blanks, or click blank then click option'

Atrace

Bload

Cscript

Dsave

Attempts:

3 left

💡 Hint

Common Mistakes

Using torch.jit.trace instead of torch.jit.script for models with control flow.

Confusing save and load methods.

torch.jit.script converts the model to TorchScript by scripting it.

3fill in blank

hard

Fix the error in the code to run inference on a mobile model.

PyTorch

input_tensor = torch.randn(1, 3, 224, 224)
output = model.[1](input_tensor)
print(output)

Drag options to blanks, or click blank then click option'

Acall

Bforward

Crun

Dpredict

Attempts:

3 left

💡 Hint

Common Mistakes

Using non-existent methods like predict or run.

Trying to call the model object directly without parentheses.

In PyTorch, the forward method runs the model on input data.

4fill in blank

hard

Fill both blanks to prepare input and run inference on a mobile model.

PyTorch

input_data = torch.randn([1])
output = model.[2](input_data)
print(output)

Drag options to blanks, or click blank then click option'

A1, 3, 224, 224

Bforward

Cpredict

D3, 224, 224

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong input shape missing batch dimension.

Using incorrect method names for inference.

Input tensor shape must match model input; use forward to run inference.

5fill in blank

hard

Fill all three blanks to save a scripted model and load it for mobile deployment.

PyTorch

scripted_model = torch.jit.[1](model)
scripted_model.[2]("model.pt")
loaded_model = torch.jit.[3]("model.pt")

Drag options to blanks, or click blank then click option'

Ascript

Bsave

Cload

Dtrace

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing up save and load methods.

Using trace instead of script for models with control flow.

Use script to convert, save to write file, and load to read file.