Emerging trends (smaller models, edge AI) in Prompt Engineering / GenAI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to define a smaller AI model for edge devices.

Prompt Engineering / GenAI

model = SmallModel(input_size=128, output_size=10, [1]=2)

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

Alayers

Bepochs

Cbatch_size

Dlearning_rate

Attempts:

3 left

💡 Hint

Common Mistakes

Confusing training parameters like epochs or batch size with model architecture.

The 'layers' parameter defines how many layers the smaller model has, which is key for edge AI.

2fill in blank

medium

Complete the code to quantize the model for efficient edge deployment.

Prompt Engineering / GenAI

quantized_model = quantize(model, [1]='int8')

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

Aprecision

Bformat

Ctype

Dmode

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'format' or 'type' which are not the correct parameter names here.

The 'precision' parameter specifies the number format, such as int8, to reduce model size.

3fill in blank

hard

Fix the error in the code to deploy the model on an edge device.

Prompt Engineering / GenAI

device = get_device('[1]')
model.to(device)

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

Agpu

Bcpu

Ctpu

Dedge

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'gpu' or 'tpu' which are not commonly available on edge devices.

Edge devices often use CPUs, so 'cpu' is the correct device string for deployment.

4fill in blank

hard

Fill both blanks to create a dictionary of model sizes and their latency on edge devices.

Prompt Engineering / GenAI

results = { [1]: [2] for [1], [2] in zip(sizes, latencies) }

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

Asize

Blatency

Csizes

Dlatencies

Attempts:

3 left

💡 Hint

Common Mistakes

Using plural variable names as keys or values, which causes errors.

The dictionary keys and values should be the singular 'size' and 'latency' from the zipped lists.

5fill in blank

hard

Fill all three blanks to filter models smaller than 10MB and with latency under 50ms.

Prompt Engineering / GenAI

filtered = { [1]: [2] for [1], [2] in models.items() if [3] < 10 and [2] < 50 }

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

Amodel_name

Bsize

Dlatency

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing up keys and values or using wrong variable names in the condition.

Keys are size, values are latency, and size is used in the condition to filter models.