Practice free →
HomeGCP ML Engineercloudcomputingmletraining › Which file format should an ML Engineer prefer f…

Which file format should an ML Engineer prefer for HIGH-VOLUME training data stored on Cloud Storage — providing efficient row-grouped reading + schema evolution for TensorFlow / PyTorch data pipelines?

ATFRecord (or Parquet for table data) — binary, compressed, splittable
BCloud DNS
CPlain text logs concatenated into a single huge file
DCloud Memorystore
Answer & Solution
Correct answer: A. TFRecord (or Parquet for table data) — binary, compressed, splittable
TFRecord / Parquet are efficient binary training formats (per PMLE §3.2). The other options aren't efficient training formats.
Solve this in the app — GCP ML Engineer practice & 24k+ MCQs →
Related questions