Transform

E457346
TFX component data preprocessing component feature engineering component

Transform is a TensorFlow Extended (TFX) component used for scalable data preprocessing and feature engineering in machine learning pipelines.

Try in SPARQL Jump to: Statements Referenced by

Statements (44)

Predicate Object
instanceOf TFX component ⓘ
data preprocessing component ⓘ
feature engineering component ⓘ
abbreviation tfx.Transform NERFINISHED ⓘ
canApply bucketization ⓘ
embedding preparation ⓘ
feature crossing ⓘ
feature scaling ⓘ
normalization ⓘ
vocabulary computation ⓘ
configuredWith preprocessing_fn ⓘ
dependsOn Apache Beam NERFINISHED ⓘ
TensorFlow NERFINISHED ⓘ
developedBy Google NERFINISHED ⓘ
documentationURL https://www.tensorflow.org/tfx/transform/get_started ⓘ
enables declarative feature engineering in TensorFlow ⓘ
preprocessing logic sharing between training and serving ⓘ
ensures preprocessing reproducibility ⓘ
training-serving skew reduction ⓘ
implementedIn Python NERFINISHED ⓘ
input ExampleGen output ⓘ
integratesWith Evaluator ⓘ
ExampleGen NERFINISHED ⓘ
Pusher NERFINISHED ⓘ
Trainer ⓘ
license Apache License 2.0 ⓘ
output metadata for transformed features ⓘ
transform graph ⓘ
transformed evaluation data ⓘ
transformed training data ⓘ
partOf TFX pipeline NERFINISHED ⓘ
TensorFlow Extended NERFINISHED ⓘ
supports TFX orchestration systems ⓘ
TensorFlow Serving NERFINISHED ⓘ
batch data processing ⓘ
distributed data processing ⓘ
large-scale datasets ⓘ
usedFor computing feature transformations ⓘ
ensuring consistent preprocessing in training and serving ⓘ
feature engineering ⓘ
preparing data for machine learning models ⓘ
scalable data preprocessing ⓘ
uses TensorFlow ops for feature transformations ⓘ
tf.Transform library NERFINISHED ⓘ

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

TensorFlow Extended → hasComponent → Transform ⓘ