Mobile Devices practice questions

A developer is building a mobile app that uses a pre-trained image classification model on-device. Which framework should they use to run the model on iOS devices?

Trap 1: Hugging Face Transformers

Hugging Face is for NLP models, not specifically for iOS image classification.

Trap 2: TensorFlow Lite

TensorFlow Lite is cross-platform but not native to iOS; Core ML is optimized for Apple hardware.

Trap 3: PyTorch Mobile

PyTorch Mobile supports Android and iOS, but Core ML is more integrated for iOS.

A
Hugging Face Transformers
Why wrong: Hugging Face is for NLP models, not specifically for iOS image classification.
B
TensorFlow Lite
Why wrong: TensorFlow Lite is cross-platform but not native to iOS; Core ML is optimized for Apple hardware.
C
PyTorch Mobile
Why wrong: PyTorch Mobile supports Android and iOS, but Core ML is more integrated for iOS.
D
Core ML
Core ML is Apple's native framework for on-device ML inference on iOS devices.

Question 2mediummultiple choice

An organisation needs to deploy PyTorch models on mobile devices with minimal latency. Which framework or tool should they use to convert and optimise the model for on-device inference?

Trap 1: TensorFlow Lite

TensorFlow Lite is for TensorFlow models; PyTorch models need a different converter.

Trap 2: Keras for mobile

Keras is a high-level API typically used with TensorFlow, not directly applicable to PyTorch models.

Trap 3: ONNX Runtime with Core ML conversion

This is possible but not the direct PyTorch path; TorchScript is the native solution.

A
TensorFlow Lite
Why wrong: TensorFlow Lite is for TensorFlow models; PyTorch models need a different converter.
B
Keras for mobile
Why wrong: Keras is a high-level API typically used with TensorFlow, not directly applicable to PyTorch models.
C
ONNX Runtime with Core ML conversion
Why wrong: This is possible but not the direct PyTorch path; TorchScript is the native solution.
D
TorchScript
TorchScript is PyTorch's own tool for model serialisation and optimisation for mobile deployment.

Question 3hardmultiple choice

An AI model is deployed to a mobile app with limited computational resources. The model is a deep neural network with high latency. Which technique is best to reduce inference time?

Trap 1: Increase batch size

Batch size affects throughput but not the latency of a single inference request.

Trap 2: Add more layers

Adding layers deepens the network and increases inference time.

Trap 3: Use a larger model

A larger model would increase computational requirements and latency.

A
Increase batch size
Why wrong: Batch size affects throughput but not the latency of a single inference request.
B
Add more layers
Why wrong: Adding layers deepens the network and increases inference time.
C
Use a larger model
Why wrong: A larger model would increase computational requirements and latency.
D
Quantization
Quantization reduces model size and speeds up inference by using lower-precision arithmetic.

Question 4mediummultiple choice

A data scientist is using a Hugging Face transformer model for a sentiment analysis task. They want to optimize inference latency for a mobile app. Which model format and framework combination is BEST suited for on-device deployment?

Trap 1: Use the full PyTorch model with JIT scripting

PyTorch JIT is not optimized for mobile; it is more suited for server inference.

Trap 2: Deploy the model on a cloud endpoint and call via REST API

Cloud endpoint adds network latency, which is undesirable for mobile app real-time inference.

Trap 3: Export to ONNX and use ONNX Runtime with GPU

ONNX Runtime with GPU is for server-side, not mobile; mobile GPUs are not standard.

A
Convert to TensorFlow Lite (TFLite) and run on the device
TFLite is optimized for mobile devices, providing low latency and small binary size.
B
Use the full PyTorch model with JIT scripting
Why wrong: PyTorch JIT is not optimized for mobile; it is more suited for server inference.
C
Deploy the model on a cloud endpoint and call via REST API
Why wrong: Cloud endpoint adds network latency, which is undesirable for mobile app real-time inference.
D
Export to ONNX and use ONNX Runtime with GPU
Why wrong: ONNX Runtime with GPU is for server-side, not mobile; mobile GPUs are not standard.

Question 5easymultiple choice

A company wants to deploy an AI model for real-time inference on edge devices with limited computational resources. Which model architecture would be MOST suitable?

Trap 1: YOLOv4

YOLOv4 is efficient but still may be too heavy for very limited edge devices; MobileNet is more optimized.

Trap 2: ResNet-152

ResNet-152 is a deep network with many parameters, too large for edge devices.

Trap 3: BERT

BERT is a large transformer model for NLP, not suitable for resource-constrained edge devices.

A
YOLOv4
Why wrong: YOLOv4 is efficient but still may be too heavy for very limited edge devices; MobileNet is more optimized.
B
MobileNet
MobileNet uses depthwise separable convolutions to reduce computation, ideal for edge deployment.
C
ResNet-152
Why wrong: ResNet-152 is a deep network with many parameters, too large for edge devices.
D
BERT
Why wrong: BERT is a large transformer model for NLP, not suitable for resource-constrained edge devices.