Why does Keras need the TensorFlow engine? I am not getting correct directions on why we need Keras. We can use TensorFlow to build a neural network model, but why do most people use Keras with TensorFlow as backend?
4$\begingroup$ Keras is just a "little helper" on top of TF. Keras comes with so many useful gadgets (e.g. early stopping, data generator, data augmentation) so that there is no need to write all the code. Also writing in TF can be a little complex. Keras is much more convenient in many situations. $\endgroup$– PeterJan 2, 2020 at 17:44
1$\begingroup$ if both can do the same task why should we club keras with tensorflow ? $\endgroup$– Aj_MLstaterJan 3, 2020 at 5:14
This makes more sense when understood in its historical context. These were the chronological events:
April 2009Theano 0.1 is released. It would dominate the deep learning framework scene for many many years.
June 2015Keras is created by François Chollet. The goal was to create an abstraction layer to make Theano easier to use, enabling fast prototyping.
August 2015Google hires François Chollet.
November 2015Tensorflow is released by Google, with much inspiration from Theano and its declarative computational graph paradigm.
December 2015Keras is refactored to allow for pluggable backend engines, and now it offers backend implementations for Theano and Tensorflow.
Other backends were later supported by Keras (CNTK, MxNet), but they never got much traction.
Time passes by and the overlap between Tensorflow and Keras grows. Tensorflow ends up duplicating many of the functionalities in Keras (apart from the multiple APIs within Tensorflow that also had big overlaps).
September 2017Theano is discontinued.
November 2017Keras is bundled with Tensorflow as
tf.keras. From this point on there are 2 different Keras: the one bundled with Tensorflow and the one that supports multiple backend engines. Both are maintained by the same people and are kept in sync at API level.
At some point, the roadmap for Tensorflow 2.0 is defined, choosing to pursue an imperative model like PyTorch. The person leading the Tensorflow API refactoring is François Chollet. This refactoring included a reorganization of the functionality to avoid duplications.
November 2018some crucial functionalities of Tensorflow are to be moved to
tf.keras, generating a heated debate
September 2019Keras 2.3 is announced to be the last release of the multi-backend version of Keras
Now, THE ANSWER to your question: Tensorflow is the most used Keras backend because it is the only one with a relevant user base that is under active development and, furthermore, the only version of Keras that is actively developed and maintained is one with Tensorflow.
So, summing up:
- At the beginning of Keras, the overlap with Tensorflow was small. Tensorflow was a bit difficult to use, and Keras simplified it a lot.
- Later, Tensorflow incorporated many functionalities similar to Keras'. Keras became less necessary.
- Then, apart from the multi-backend version, Keras was bundled with Tensorflow. Their separation line blurred over the years.
- The multi-backend Keras version was discontinued. Now the only Keras is the one bundled with Tensorflow.
Update: the relationship between Keras and Tensorflow is best understood with an example:
The dependency between Keras and Tensorflow is internal to Keras, it is not exposed to the programmer working with Keras. For example, in the source code of Keras, there is an implementation of a convolutional layer; this implementation calls package
keras.backend to actually run the convolution computation; depending on the Keras configuration file, this backend is set to use the Tensorflow backend implementation in
keras.backend.tensorflow_backend.py; this Keras file just invokes Tensorflow to compute the convolution
Update 2: new important events in the timeline:
August 2021: Tensorflow 2.6.0 no longer has Keras as part of it. Keras has now its own PIP package (
keras) and lives on its own github repo.
1$\begingroup$ i can write a simple neural network with keras . I can write simple neural network in tensoflow as well .when the same task can be done on both why do we need to club tensorflow with keras ? i found from many blog post that keras is API and Tensorflow is framework ,can you tell me how exactly this make difference why keras cannot stand on its own ? $\endgroup$ Jan 3, 2020 at 5:02
4$\begingroup$ Keras was built on top of Tensorflow: Keras was a library that used Tensorflow to work, but providing a simpler interface. But then, Keras itself was incorporated into Tensorflow and later they became the same thing. Many blog posts are from the time were Keras still existed on its own, but that's not the case anymore. $\endgroup$– noeJan 3, 2020 at 8:54
$\begingroup$ @Aj_MLstater If you write a simple neural network with Keras, then it does not implement the core calculation operations directly, it uses tensorflow (or some other backend) constructs and code to run that simple neural network. Keras cannot stand on its own because it intentionally does not contain all the code that would implement 'standing on its own'. $\endgroup$– PeterisJan 3, 2020 at 12:05
$\begingroup$ @peteris thanks for the information , to build a simple neural network with Keras it needs tensorflow to do the calculation but the simple neural network can be written directly in tensorflow itself right ? what we want to use keras in the first place and take the hardship of integrating it with tensorflow . what makes keras so cool that rather than using tensorflow directly people are integrating keras with tensorflow ?can you please explain with a sample code that will help everyone to understand the benfit of using tensorflow vs keras+tensorflow ? $\endgroup$ Jan 3, 2020 at 13:10
$\begingroup$ @ncasas ,thanks for the information . can you share a small example which show on why the dependency is needed between keras and tensorflow is needed ? can we consider keras like a scikit learn library ? , can you direct me to any link or share a short code with example that will help everyone to understand these 2 points 1.what is lagging in tensorflow that it needs keras or 2. what is lagging in keras that it needs tensorflow ? or benfit of using tensorflow vs keras+tensorflow . $\endgroup$ Jan 3, 2020 at 13:12
Keras is an application programming interface (API). It is a single interface that can support multi-backends, which means a programmer can write Keras code once and it can be executed in a variety of neural networks frameworks (e.g., TensorFlow, CNTK, or Theano).
TensorFlow 2.0 is the suggested backend starting with Keras 2.3.0.
1$\begingroup$ i found from many blog post that keras is API and Tensorflow is framework ,can you tell me how exactly this make difference ,what is the difference between being a API and framework ? why keras cannot stand on its own because its a API? $\endgroup$ Jan 3, 2020 at 5:17
1$\begingroup$ API is an interface. Think of C++ header files. TF is a framework ("bundle of stuff") which has an implementation of the Keras API (C++ source files). However Keras (the package which has multiple backends) is in itself an implementation of the Keras API. The "implementation" here is: "Use this part of a backend to run the function/layer". You might also compare "API" to "Standard": Some specification on how things work. $\endgroup$ Jan 5, 2020 at 15:04
Lets go back to basics here.
It is not possible to only use Keras without using a backend, such as Tensorflow, because Keras is only an extension for making it easier to read and write machine learning programs. All the actual calculations needed to create models are not implemented in Keras, which is why you need to use a backend library for anything to work.
When you are creating a model in Keras, you are actually still creating a model using Tensorflow, Keras just makes it easier to code.
1$\begingroup$ thanks for the information . keras makes easier to write machine learning program ,how exactly dose it make difference in writing the same code in tensorflow can you share a sample code or direct me to any link to know more about it and tensorflow dose not need any backend engine right so why everyone are using keras+tensorflow rather than using tensorflow directly ? is tensorflow hard for coding to build model ? $\endgroup$ Jan 3, 2020 at 13:18
1$\begingroup$ And now we have the ability to convert Keras models into Tensorflow estimators...tensorflow.org/guide/… $\endgroup$ Jun 15, 2021 at 1:53
Additionally: Think of it as an abstraction layer.
Keras gives nice and intuitive way to build and think about neural network, but you have to understand thats not how computer takes orders. Hiding this complexity behind Tensorflow allows us to think naturally about building a neural network and not all the details behind implementation.
(On a general note thats why python is so popular, cause it abstracts the complexity away, and allows you to think and write down solution more naturally and intuitively)
$\begingroup$ i found from many blog post that keras is API and Tensorflow is framework ,can you tell me how exactly this make difference ,what is the difference between being a API and framework ? why keras cannot stand on its own because its a API? can you please tell more about how this abstraction work for keras and dose tensorflow have a abstraction as well ? because i can write a simple neural network with keras . I can write simple neural network in tensoflow as well . abstraction hide the complexity so is keras a programming language like pyhton ? $\endgroup$ Jan 3, 2020 at 5:18
The first point to note is that Keras can potentially use many backends (e.g. Theano before it was discontinued, Microsoft Cognitive Toolkit, to name a couple). It just so happens that Keras has proven to be the most popular among the community. As a result, TensorFlow has adapted to the extent that Keras is now the default API in TensorFlow 2.0.
One of the biggest changes is the way libraries are now loaded using tf.keras.
Consider this example. Let's say one wishes to run a Sequential model using Keras. To do so, one must import the relevant libraries.
In the first version of TensorFlow, it would be done as follows:
from tensorflow.python.keras.models import Sequential from tensorflow.python.keras.layers import Dense from tensorflow.python.keras.wrappers.scikit_learn import KerasRegressor
The model is defined as such:
model = Sequential() model.add(Dense(8, activation='relu', input_shape=(4,))) model.add(Dense(1, activation='sigmoid'))
Now, let's contrast this to the TensorFlow 2.0 notation:
from tensorflow.keras import models from tensorflow.keras import layers model = models.Sequential() model.add(layers.Dense(8, activation='relu', input_shape=(4,))) model.add(layers.Dense(1, activation='sigmoid'))
The Sequential function is now being defined using models, and layers is the only other library imported. Whereas in TensorFlow v1.0, Sequential, Dense, and KerasRegressor all had to be imported separately to generate the model.
Using the above example as a reference point, one can say that Keras now uses TensorFlow as a backend most frequently - simply because it has proven to be the most popular. As a result, TensorFlow has adapted to making the syntax for calling Keras more user-friendly, and thus Keras has become the default API in v2.0.
You might also find this article of use for further information on this topic: https://www.pyimagesearch.com/2019/10/21/keras-vs-tf-keras-whats-the-difference-in-tensorflow-2-0/
1$\begingroup$ i can write a simple neural network with keras . I can write simple neural network in tensoflow as well .when the same task can be done on both why do we need to club tensorflow with keras ? i found from many blog post that keras is API and Tensorflow is framework ,can you tell me how exactly this make difference why keras cannot stand on its own ? why excatly keras need a backend and why tensorflow dose not need a backend ? $\endgroup$ Jan 3, 2020 at 5:06
Keras used to use 2 backends(Theano and Tensorflow), but now only supports Tensorflow because of the discontinuation of Theano. The reason why Keras uses Tensorflow as it's backend is because it is an abstraction layer.
It is the easiest way to get started with AI and machine learning because all of the core algorithms are implemented in tensorflow and keras allows you to just call the classes/functions without adding any additional code. Great starter library for beginners and AI enthusiasts who have little coding experience.
1$\begingroup$ why keras needs a backend ?why tensorflow dose not need a backend ? $\endgroup$ Jan 3, 2020 at 5:05
2$\begingroup$ Thats a good question @Aj_MLstater. Tensorflow does not need a backend because everything that is built using tensorflow i.e, the supervised, unsupervised algorithms are built from scratch and keras is a library which uses these algorithms that are built in Tensorflow as a backend and makes it easier for the developers to get the results easily without have an immense knowledge about the algorithm. I hope this helps. Cheers $\endgroup$ Jan 6, 2020 at 2:49
Imagine you have a basic maths framework, a lot of functions doing addition, subtraction, multiplication and division.
Imagine in everyday life you often need to compute averages.
Then you make a function (using the functions from the framework, inside it), that will take an array of numbers as parameters an return the mean.
The framework is actually doing the work, it's still a lot of additions and a division, but your API-like function is a way nicer way to do what you need.
Let's say you were using Numpy (an algebra framework on CPU) to do your stuff. Numba is it's equivalent but on GPU. If in your code you had a lot of "numpy.add(a, b)" everywhere you needed an addition, you would need to change it everywhere to "numba.add(a, b)", so a lot of shitty work. But if instead you were using your homemade function "add(a, b)", then you just have to change the framework you use inside your function, easy peasy ! So yeah you understood correctly, it's better to update the API than the framework. To come back to this dumb example, Numpy is a "CPU computing framework", so it wouldn't make any sense to change it to use the GPU (Numba was created for it). But your custom function can easily be modified, as it's purpose is to "do the job the best way for you". So it's good practice to stick to using your "API" everywhere, even if it sometimes seems unnecessary.
Now picture Keras as this function, and Tensorflow as the algebra framework. Sure, most of the time you can use directly the framework, but if you want your code to be cleaner, you'll use your API.
As of today Keras and Tensorflow are bundled together and Tensorflow interface is getting closer to it, but that was the idea.
If you can do the same model as easily with Keras than directly from Tensorflow, it might seems better to get rid of the "useless" middle-man (Keras), but beware! If one day Tensorflow implements a better way of doing it, Keras will use it, while using directly Tensorflow you'll need to update your code...
I over simplify everything, I know, but you seems to have a hard time to distinguish framework and API.
You can see that the API is dumb, meaning that the API on itself is using the algebra framework and would be useless without it. Or it would need to implement all those operation and become a fully-fledged framework instead of a simple API. The API needs the framework to work, as Keras needs TensorFlow.
1$\begingroup$ thanks but i did not understand these 2 lines " use Numba to do these computations on GPU, you will just need to update your custom function instead of your whole code." --> dose it mean we need to update API rather than Framework or vice versa ? and in last 3rd paragraph of your answer why gid rid of keras when you said it is easy ? thensorflow implemetns better way of doing it thne why keras shoudl use it ? , what dose "while using directly Tensorflow you'll need to update your code... " actual mean ? can you explain more about this latst 3rd para of ur answer ? $\endgroup$ Jan 4, 2020 at 22:26
2$\begingroup$ I updated a bit. TensorFlow is a framework (meaning it does stuff with a lot of advanced functions) and have his own API (the way you use it, user-friendly functions). Keras is just another API (other even more user-friendly functions). Keras is meant to be a THE stable API. It means that in future versions of TF (see TF 2.0 for instance), the way to use it can change quite a lot. So if you use TF directly, you might need to change a lot of stuff when there is a big update, while using Keras you should not need it, or way less (Keras's guys would do it for you, inside Keras functions) $\endgroup$ Jan 6, 2020 at 0:51