Shulou(Shulou.com)06/01 Report--

What this article shares with you is the analysis of how to carry out Keras. The editor thinks it is very practical, so I share it with you to learn. I hope you can get something after reading this article.

Introduction to Keras

Keras is an advanced neural network API written in Python, which can run with TensorFlow, CNTK, or Theano as the back end. The development focus of Keras is to support rapid experimentation. The key to good research is to be able to convert your ideas into experimental results with minimal delay.

Keras usage scenario

Allows simple and rapid prototyping (due to user-friendly, highly modular, scalable).

At the same time, convolution neural network and cyclic neural network are supported, as well as the combination of both.

Runs seamlessly on CPU and GPU.

Characteristics of Keras

User-friendly. Keras is an API designed for humans, not machines. It puts user experience at the top and center. Keras follows best practices to reduce cognitive difficulties: it provides a consistent and simple API, minimizes the number of user actions required for common use cases, and provides clear and actionable feedback when users make mistakes.

Modularization. The model is understood as a sequence or diagram composed of independent and fully configurable modules. These modules can be assembled with as few restrictions as possible. In particular, the neural network layer, loss function, optimizer, initialization method, activation function and regularization method are all modules that can be combined to build a new model.

Easy to scale. New modules are easy to add (as new classes and functions), and existing modules provide plenty of examples. Because it is easy to create new modules that can improve performance, Keras is more suitable for advanced research.

Based on Python implementation. Keras does not have a separate configuration file in a specific format. The model is defined in Python code, which is compact, easy to debug, and easy to extend.

Get started with Keras

The core data structure of Keras is model, a way to organize the network layer. The simplest model is the Sequential sequential model, which is linearly stacked by multiple network layers. For more complex structures, you should use the Keras functional API, which allows arbitrary neural network diagrams to be constructed. The Sequential model is as follows:

From keras.models import Sequentialmodel = Sequential ()

You can simply use .add () to stack the model:

From keras.layers import Densemodel.add (Dense (units=64, activation='relu', input_dim=100)) model.add (Dense (units=10, activation='softmax'))

After you have finished building the model, you can use .compile () to configure the learning process:

Model.compile (loss='categorical_crossentropy',optimizer='sgd',metrics= ['accuracy'])

If necessary, you can further configure your optimizer. The core principle of Keras is to make things fairly simple while allowing users to have complete control when needed (the ultimate control is the extensibility of the source code).

Model.compile (loss=keras.losses.categorical_crossentropy,optimizer=keras.optimizers.SGD (lr=0.01, momentum=0.9, nesterov=True))

Now you can iterate over the training data in batches:

# x_train and y_train are Numpy arrays-- just like in Scikit-Learn API. Model.fit (x_train, y_train, epochs=5, batch_size=32)

Alternatively, you can manually provide the batch data to the model:

Model.train_on_batch (x_batch, y_batch)

It only takes one line of code to evaluate the performance of the model:

Loss_and_metrics = model.evaluate (x_test, y_test, batch_size=128)

Or generate predictions for new data:

Classes = model.predict (x_test, batch_size=128) the above is how to analyze Keras. The editor believes that there are some knowledge points that we may see or use in our daily work. I hope you can learn more from this article. For more details, please follow the industry information channel.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.