Download as PDF, PPTX
![Compile model # Compile model model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy']) ● The loss function to use to evaluate a set of weights. ● The optimizer used to search through different weights for the network and any optional metrics we would like to collect and report during training. ● For classification problem you will want to set this to metrics=[‘accuracy’]](https://image.slidesharecdn.com/kerasfordeeplearning-170611231806/75/Keras-Deep-Learning-Library-for-Python-6-2048.jpg)
![Evaluate Model score = model.evaluate(x_test, y_test, verbose=0) print('Test loss:', score[0]) print('Test accuracy:', score[1]) ● It give us an idea of how well we have modeled the dataset.](https://image.slidesharecdn.com/kerasfordeeplearning-170611231806/75/Keras-Deep-Learning-Library-for-Python-8-2048.jpg)
Uploaded byRafi Khan
Keras: Deep Learning Library for Python
Keras is a high-level neural networks API in Python, built on Theano or TensorFlow, known for its user-friendliness and support for various types of neural networks. It allows for easy model creation, including defining architecture, compiling with loss functions, fitting with training data, and making predictions. The document provides example code for creating a convolutional model and includes details on training and evaluating model accuracy.
In this document
Powered by AI
Overview of Keras as a high-level neural networks API in Python used for deep learning.
Steps to create a Keras model, including architecture, compilation, fitting, evaluation, and prediction.
A brief thank you note.
Keras: Deep Learning Library for Python
- 1. Keras A deep learninglibrary
- 2. What is keras? ●Keras is a high-level neural networks API, written in Python. ● Built on top of either Theano or TensorFlow. ● Most powerful & easy to use for developing and evaluating deep learning models.
- 3. Why use Keras? ●Allows for easy and fast prototyping (through user friendliness, modularity, and extensibility). ● Supports both convolutional networks and recurrent networks, as well as combinations of the two. ● Runs seamlessly on CPU and GPU.
- 4. Creating a kerasmodel ● Architecture Definition:-no of layers,no of nodes in layers,activation function to be used. ● Compile:-defines the loss function and some details about how optimization works. ● Fit:-cycle of backpropagation and optimization of model weights with your data. ● Predict:-to predict the model prepared.
- 5. Keras code forcreating model The sequential model used is a linear stack of layers. ### Model begins ### model = Sequential() model.add(Convolution2D(16, 5, 5, activation='relu', input_shape=(img_width, img_height, 3))) model.add(MaxPooling2D(2, 2)) model.add(Convolution2D(32, 5, 5, activation='relu')) model.add(MaxPooling2D(2, 2)) model.add(Flatten()) model.add(Dense(1000, activation='relu')) model.add(Dense(10, activation='softmax')) ### Model Ends ###
- 6. Compile model # Compilemodel model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy']) ● The loss function to use to evaluate a set of weights. ● The optimizer used to search through different weights for the network and any optional metrics we would like to collect and report during training. ● For classification problem you will want to set this to metrics=[‘accuracy’]
- 7. Fit the model #fit model model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs, verbose=1, validation_data=(x_test, y_test)) ● Execution of model for some data. ● Train data and iterate data in batches.
- 8. Evaluate Model score =model.evaluate(x_test, y_test, verbose=0) print('Test loss:', score[0]) print('Test accuracy:', score[1]) ● It give us an idea of how well we have modeled the dataset.
- 9. Predict classes=model.predict(x_test,batch_size=128) ● It generatesprediction on new data
- 10.