Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Remove unnecessary words and organize content in keras.md #2311

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Remove unnecessary words and organize content in keras.md #2311

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
416a884
Update keras.md
stumblefiend May 22, 2024
4c17377
Remove non-existent link to researchers intro
stumblefiend May 24, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
140 changes: 66 additions & 74 deletions site/en/guide/keras.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,115 +1,107 @@
# Keras: The high-level API for TensorFlow

Keras is the high-level API of the TensorFlow platform. It provides an
approachable, highly-productive interface for solving machine learning (ML)
problems, with a focus on modern deep learning. Keras covers every step of the
machine learning workflow, from data processing to hyperparameter tuning to
deployment. It was developed with a focus on enabling fast experimentation.

With Keras, you have full access to the scalability and cross-platform
capabilities of TensorFlow. You can run Keras on a TPU Pod or large clusters of
GPUs, and you can export Keras models to run in the browser or on mobile
devices. You can also serve Keras models via a web API.

Keras is designed to reduce cognitive load by achieving the following goals:

* Offer simple, consistent interfaces.
* Minimize the number of actions required for common use cases.
* Provide clear, actionable error messages.
* Follow the principle of progressive disclosure of complexity: It's easy to get
started, and you can complete advanced workflows by learning as you go.
* Help you write concise, readable code.

## Who should use Keras

The short answer is that every TensorFlow user should use the Keras APIs by
default. Whether you're an engineer, a researcher, or an ML practitioner, you
should start with Keras.

There are a few use cases (for example, building tools on top of TensorFlow or
developing your own high-performance platform) that require the low-level
[TensorFlow Core APIs](https://www.tensorflow.org/guide/core). But if your use
case doesn't fall into one
of the
[Core API applications](https://www.tensorflow.org/guide/core#core_api_applications),
you should prefer Keras.
Keras is the approachable, productive, high-level API for fast
experimentation on the TensorFlow platform. Keras focuses on modern
deep learning and lets you fully use TensorFlow's scalability and
cross-platform features.

If you use TensorFlow, use Keras APIs by default for
each step of the machine learning workflow, from data processing to
hyperparameter tuning to deployment. Otherwise,
[a few use cases](https://www.tensorflow.org/guide/core#who_should_use_the_core_apis)
require the low-level [TensorFlow Core APIs](https://www.tensorflow.org/guide/core).

There are multiple ways to run a Keras model:

* Run on a TPU Pod or large clusters of GPUs.
* Export to run in a browser or mobile device.
* Serve via a web API.

Keras makes ML workflows easier:

* Simple, consistent interfaces.
* Few steps for common tasks.
* Clear, helpful error messages.
* Gradual learning curve: easy to start, advanced options later.
* Concise, readable code.

## Keras API components

The core data structures of Keras are [layers](https://keras.io/api/layers/) and
[models](https://keras.io/api/models/). A layer is a simple input/output
transformation, and a model is a directed acyclic graph (DAG) of layers.
Keras has two core data structures:

* [Layers](https://keras.io/api/layers/) - A simple input/output
transformation.
* [Models](https://keras.io/api/models/) - A directed acyclic graph (DAG) of layers.

### Layers

The `tf.keras.layers.Layer` class is the fundamental abstraction in Keras. A
`Layer` encapsulates a state (weights) and some computation (defined in the
`tf.keras.layers.Layer.call` method).
The `tf.keras.layers.Layer` class is the main abstraction in Keras. A
`Layer` encapsulates a state (weights) and some computation defined in the
`tf.keras.layers.Layer.call` method.

Weights created by layers can be trainable or non-trainable. Layers are
recursively composable: If you assign a layer instance as an attribute of
another layer, the outer layer will start tracking the weights created by the
Weights created by layers are trainable or non-trainable. Layers are
recursively composable: for a layer instance assigned as an attribute of
another layer, the outer layer tracks the weights created by the
inner layer.

You can also use layers to handle data preprocessing tasks like normalization
and text vectorization. Preprocessing layers can be included directly into a
model, either during or after training, which makes the model portable.
Layers can also handle data preprocessing tasks like normalization
and text vectorization. Models are made portable by directly including
preprocessing layers during or after training.

### Models

A model is an object that groups layers together and that can be trained on
data.
A model is an object that groups layers and trains on data. There are a few
main model types:

* [`Sequential` model](https://www.tensorflow.org/guide/keras/sequential_model) - The
simplest model which is a linear stack of layers.
* [Keras functional API](https://www.tensorflow.org/guide/keras/functional_api) - Lets
you build arbitrary graphs of layers for more complex architectures.

The simplest type of model is the
[`Sequential` model](https://www.tensorflow.org/guide/keras/sequential_model),
which is a linear stack of layers. For more complex architectures, you can
either use the
[Keras functional API](https://www.tensorflow.org/guide/keras/functional_api),
which lets you build arbitrary graphs of layers, or
[use subclassing to write models from scratch](https://www.tensorflow.org/guide/keras/making_new_layers_and_models_via_subclassing).
You can also [use subclassing to write models from scratch](https://www.tensorflow.org/guide/keras/making_new_layers_and_models_via_subclassing).

The `tf.keras.Model` class features built-in training and evaluation methods:
The `tf.keras.Model` class has built-in training and evaluation methods:

* `tf.keras.Model.fit`: Trains the model for a fixed number of epochs.
* `tf.keras.Model.predict`: Generates output predictions for the input samples.
* `tf.keras.Model.evaluate`: Returns the loss and metrics values for the model;
* `tf.keras.Model.fit` - Trains the model for a fixed number of epochs.
* `tf.keras.Model.predict` - Generates output predictions for the input samples.
* `tf.keras.Model.evaluate` - Returns the loss and metrics values for the model and is
configured via the `tf.keras.Model.compile` method.

These methods give you access to the following built-in training features:
These methods provide built-in training features:

* [Callbacks](https://www.tensorflow.org/api_docs/python/tf/keras/callbacks).
You can leverage built-in callbacks for early stopping, model checkpointing,
* [Callbacks](https://www.tensorflow.org/api_docs/python/tf/keras/callbacks) -
Leverage built-in callbacks for early stopping, model checkpointing,
and [TensorBoard](https://www.tensorflow.org/tensorboard) monitoring. You can
also
[implement custom callbacks](https://www.tensorflow.org/guide/keras/writing_your_own_callbacks).
* [Distributed training](https://www.tensorflow.org/guide/keras/distributed_training).
You can easily scale up your training to multiple GPUs, TPUs, or devices.
* Step fusing. With the `steps_per_execution` argument in
`tf.keras.Model.compile`, you can process multiple batches in a single
`tf.function` call, which greatly improves device utilization on TPUs.
* [Distributed training](https://www.tensorflow.org/guide/keras/distributed_training) -
Easily scale training to multiple GPUs, TPUs, or devices.
* Step fusing - The `steps_per_execution` argument in
`tf.keras.Model.compile` can process multiple batches in a single
`tf.function` call, improving device utilization on TPUs.

For a detailed overview of how to use `fit`, see the
For details on using `fit`, see the
[training and evaluation guide](https://www.tensorflow.org/guide/keras/training_with_built_in_methods).
To learn how to customize the built-in training and evaluation loops, see
[Customizing what happens in `fit()`](https://www.tensorflow.org/guide/keras/customizing_what_happens_in_fit).
[customizing what happens in `fit()`](https://www.tensorflow.org/guide/keras/customizing_what_happens_in_fit).

### Other APIs and tools

Keras provides many other APIs and tools for deep learning, including:
Keras provides many other APIs and tools for deep learning:

* [Optimizers](https://keras.io/api/optimizers/)
* [Metrics](https://keras.io/api/metrics/)
* [Losses](https://keras.io/api/losses/)
* [Data loading utilities](https://keras.io/api/data_loading/)

For a full list of available APIs, see the
[Keras API reference](https://keras.io/api/). To learn more about other Keras
projects and initiatives, see
[The Keras ecosystem](https://keras.io/getting_started/ecosystem/).
[Keras API reference](https://keras.io/api/). See
[the Keras ecosystem](https://keras.io/getting_started/ecosystem/) to learn more
about other Keras projects.

## Next steps

To get started using Keras with TensorFlow, check out the following topics:
To get started using Keras with TensorFlow, see the following topics:

* [The Sequential model](https://www.tensorflow.org/guide/keras/sequential_model)
* [The Functional API](https://www.tensorflow.org/guide/keras/functional)
Expand All @@ -132,4 +124,4 @@ To learn more about Keras, see the following topics at
* [Introduction to Keras for Engineers](https://keras.io/getting_started/intro_to_keras_for_engineers/)
* [Introduction to Keras for Researchers](https://keras.io/getting_started/intro_to_keras_for_researchers/)
stumblefiend marked this conversation as resolved.
Show resolved Hide resolved
stumblefiend marked this conversation as resolved.
Show resolved Hide resolved
* [Keras API reference](https://keras.io/api/)
* [The Keras ecosystem](https://keras.io/getting_started/ecosystem/)
* [The Keras ecosystem](https://keras.io/getting_started/ecosystem/)