TensorFlow Vision MCQ 15 Questions
Time: ~25 mins Intermediate

TensorFlow Vision MCQ

tf.data pipelines, `keras.applications`, and preprocessing layers integrated in the TF/Keras 2 workflow.

Easy: 5 Q Medium: 6 Q Hard: 4 Q
TF / Keras

Stack

tf.data

Pipeline

Applications

Xception…

SavedModel

Serve

TensorFlow for vision

Use tf.data.Dataset to build efficient input pipelines (map, batch, prefetch). keras.applications provides pretrained CNNs; preprocessing can live in the model via Rescaling / RandomFlip layers for export-friendly graphs. SavedModel packages inference for TF Serving and mobile converters.

Augment in the graph

Keras preprocessing layers keep train and serve transforms aligned when exported.

Key ideas

tf.data

from_tensor_slices, map, batch, prefetch to device.

Rescaling / Augment

Layers for resize, flip, rotate in-model.

applications

ResNet50(include_top=False) feature extractor.

SavedModel

Signatures for serving and TFLite conversion.

Fine-tune pattern

base = Xception(weights='imagenet', include_top=False) → GlobalAveragePooling2D → Dense

Pro tip: Use mixed precision policy on supported GPUs for faster training with minimal code changes.