Keras

E18356

Keras is a high-level neural networks API written in Python that simplifies building, training, and deploying deep learning models, often running on top of frameworks like TensorFlow.

Observed surface forms (3)

Surface form Occurrences
Keras team 1
TensorFlow Keras API 1
tf.keras 1

Statements (52)

Predicate Object
instanceOf deep learning library
neural network API
software library
aim extensibility
fast experimentation
modularity
user-friendliness
developer François Chollet
feature batch normalization layers
callback system
data preprocessing utilities
dropout layers
embedding layers
high-level abstraction for neural networks
layer-based model definition
metrics and loss functions
model serialization
model subclassing API
optimizers
pretrained models
regularization techniques
firstReleaseYear 2015
genre high-level API
machine learning framework
neural network framework
integratedInto TensorFlow
license MIT License
maintainer Keras self-linksurface differs
surface form: Keras team
operatingSystem cross-platform
partOf TensorFlow
surface form: TensorFlow 2.x core API
primaryBackend TensorFlow
programmingLanguage Python
repository https://github.com/keras-team/keras
supports CPU execution
GPU acceleration via backend
autoencoders
convolutional neural networks
fully connected networks
functional API for complex models
generative adversarial networks
recurrent neural networks
sequential API for simple models
supportsBackend Microsoft Cognitive Toolkit
PlaidML
TensorFlow
Theano
useCase deep learning model development
model deployment
neural network evaluation
neural network training
website https://keras.io/
writtenIn Python

Referenced by (10)

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

this entity surface form: TensorFlow Keras API
TensorFlow hasComponent Keras
this entity surface form: tf.keras
Keras maintainer Keras self-linksurface differs
this entity surface form: Keras team