torchvision detection models in PyTorch - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Model Choice

intermediate

Choosing the right torchvision detection model for small objects

You want to detect very small objects in images using torchvision detection models. Which model is generally best suited for this task?

A RetinaNet without FPN

B SSD with a VGG16 backbone

C Faster R-CNN with a ResNet-50 backbone and Feature Pyramid Network (FPN)

D Mask R-CNN without any backbone network

💡 Hint

Think about models that use multi-scale features to detect objects of different sizes.

Faster R-CNN with FPN uses multi-scale feature maps, which helps detect small objects better than single-scale models like SSD or RetinaNet without FPN. Mask R-CNN without a backbone is invalid.

❓ Predict Output

intermediate

Output shape of torchvision Faster R-CNN predictions

Given a batch of 2 images passed through a pretrained Faster R-CNN model from torchvision, what is the type and length of the output?

PyTorch

import torch
import torchvision
model = torchvision.models.detection.fasterrcnn_resnet50_fpn(pretrained=True)
model.eval()
inputs = [torch.randn(3, 300, 400), torch.randn(3, 500, 600)]
outputs = model(inputs)
print(type(outputs), len(outputs))

A <class 'list'> 2

B <class 'torch.Tensor'> 2

C <class 'dict'> 2

D <class 'list'> 1

💡 Hint

Check the model's documentation for output format when passing a list of images.

torchvision detection models return a list of dictionaries, one per input image. So the output is a list with length equal to batch size.

❓ Hyperparameter

advanced

Effect of changing the IoU threshold in torchvision detection models

In torchvision's Faster R-CNN, what is the effect of increasing the IoU threshold used in the Non-Maximum Suppression (NMS) step during inference?

A More overlapping boxes are kept, potentially increasing duplicate detections

B Fewer boxes are kept, reducing duplicate detections but possibly missing objects

C The model trains faster but with lower accuracy

D The backbone network changes its feature extraction layers

💡 Hint

Think about what happens when the threshold for overlap to suppress boxes is higher.

Increasing the IoU threshold means boxes must overlap more to be suppressed, so more boxes remain, which can cause duplicates.

🔧 Debug

advanced

Debugging a runtime error in torchvision Mask R-CNN training

You get the error: 'TypeError: forward() missing 1 required positional argument: "targets"' when training Mask R-CNN from torchvision. What is the most likely cause?

A The input images are not normalized correctly

B You forgot to pass the target annotations (bounding boxes and masks) to the model during training

C The model backbone is not pretrained

D You passed the targets as a list instead of a dictionary

💡 Hint

Check the model's forward method signature for training mode.

During training, torchvision detection models require both images and targets. Missing targets causes this error.

❓ Metrics

expert

Interpreting mAP metric for torchvision detection models

You evaluate a torchvision Faster R-CNN model on a test set and get a mean Average Precision (mAP) of 0.75 at IoU=0.5. What does this number mean?

A The model detects 75% of objects but with no overlap requirement

B The model's classification accuracy is 75% on detected objects

C The model's loss during training was 0.75 at IoU=0.5

D On average, the model correctly detects 75% of objects with at least 50% overlap with ground truth

💡 Hint

Recall what mAP and IoU thresholds represent in object detection evaluation.

mAP at IoU=0.5 means the average precision of detections that overlap ground truth by at least 50%. A value of 0.75 means good detection quality.

Practice

1. Which of the following is a detection model provided by torchvision for finding objects in images?

easy

A. ResNet

B. Faster R-CNN

C. VGG16

D. Transformer

Solution

1. Step 1: Understand torchvision detection models

   torchvision offers models specifically for object detection like Faster R-CNN, Mask R-CNN, and RetinaNet.

2. Step 2: Identify the correct model

   ResNet and VGG16 are classification models, and Transformer is a general architecture, not a torchvision detection model.

3. Final Answer:

   Faster R-CNN -> Option B

4. Quick Check:

   Detection model = Faster R-CNN [OK]

Hint: Detection models include 'R-CNN' names in torchvision [OK]

Common Mistakes:

• Confusing classification models with detection models
• Thinking Transformer is a torchvision detection model
• Mixing up model types

2. Which code snippet correctly loads a pretrained Faster R-CNN model from torchvision?

easy

A. model = torchvision.models.detection.fasterrcnn_resnet50_fpn(weights='DEFAULT')

B. model = torchvision.models.fasterrcnn_resnet50_fpn(pretrained=True)

C. model = torchvision.detection.fasterrcnn_resnet50_fpn(pretrained=True)

D. model = torchvision.models.detection.fasterrcnn_resnet50_fpn(pretrained=False)

Solution

1. Step 1: Check the correct module and argument

   In recent torchvision versions, detection models are under torchvision.models.detection and use weights='DEFAULT' instead of pretrained=True.

2. Step 2: Identify the correct syntax

   model = torchvision.models.detection.fasterrcnn_resnet50_fpn(weights='DEFAULT') uses the correct module path and argument for loading pretrained weights.

3. Final Answer:

   model = torchvision.models.detection.fasterrcnn_resnet50_fpn(weights='DEFAULT') -> Option A

4. Quick Check:

   Use weights='DEFAULT' to load pretrained detection models [OK]

Hint: Use weights='DEFAULT' for pretrained detection models [OK]

Common Mistakes:

• Using pretrained=True instead of weights='DEFAULT'
• Wrong module path
• Missing 'models' in import path

3. What will be the output type of the following code snippet?

import torch
import torchvision
model = torchvision.models.detection.fasterrcnn_resnet50_fpn(weights='DEFAULT')
model.eval()
inputs = [torch.rand(3, 300, 400)]
outputs = model(inputs)
print(type(outputs))

medium

A.

B.

C.

D.

Solution

1. Step 1: Understand model output format

   torchvision detection models return a list of dictionaries, one per input image.

2. Step 2: Check the type of outputs variable

   Since inputs is a list with one tensor, outputs will be a list containing one dictionary.

3. Final Answer:

   <class 'list'> -> Option D

4. Quick Check:

   Detection model output = list of dicts [OK]

Hint: Detection models output a list, one dict per input image [OK]

Common Mistakes:

• Expecting a single tensor output
• Confusing output with dict directly
• Not wrapping inputs in a list

4. Identify the error in this code snippet that tries to use a pretrained Mask R-CNN model:

import torchvision
model = torchvision.models.detection.maskrcnn_resnet50_fpn(pretrained=True)
model.eval()

medium

A. The function name should be fasterrcnn_resnet50_fpn

B. Mask R-CNN is not available in torchvision

C. The argument 'pretrained' is deprecated; use 'weights' instead

D. Missing import for torch

Solution

1. Step 1: Check function and argument usage

   Mask R-CNN is available as maskrcnn_resnet50_fpn in torchvision.models.detection.

2. Step 2: Identify deprecated argument

   Using pretrained=True is deprecated; the correct way is weights='DEFAULT'.

3. Final Answer:

   The argument 'pretrained' is deprecated; use 'weights' instead -> Option C

4. Quick Check:

   Use weights='DEFAULT' instead of pretrained=True [OK]

Hint: Replace pretrained=True with weights='DEFAULT' [OK]

Common Mistakes:

• Using deprecated pretrained argument
• Confusing model names
• Assuming Mask R-CNN is missing

5. You want to fine-tune a pretrained RetinaNet model from torchvision on your custom dataset with 5 classes. Which step is necessary before training?

hard

A. Replace the model's box predictor head to output 6 classes (5 + background)

B. Change the input image size to 224x224

C. Set pretrained=False to train from scratch

D. Freeze all layers except the backbone

Solution

1. Step 1: Understand class output adjustment

   Detection models include a background class, so for 5 classes, output head must predict 6 classes.

2. Step 2: Modify the box predictor head

   Replace the box predictor to match the number of classes (5 + 1 background) before training.

3. Final Answer:

   Replace the model's box predictor head to output 6 classes (5 + background) -> Option A

4. Quick Check:

   Adjust output head for custom classes [OK]

Hint: Add 1 for background class when changing output head [OK]

Common Mistakes:

• Not adjusting output head for new classes
• Changing input size unnecessarily
• Freezing all layers without fine-tuning head