Computer vision with Keras and R
1
Presentation
2
Foundamental of machine learning
2.1
Types
2.2
Model performance evaluation
2.2.1
Training, validation and test sets
2.2.2
Evaluation metrics
3
Neural Networks
3.1
Structure of neural network
3.1.1
Tensors
3.1.2
Layers
3.1.3
Activation functions
3.1.4
Loss functions and optimizers
3.1.5
Building a Neural Network from scratch
3.2
Introduction to Keras
3.2.1
Installing keras
3.2.2
building model with keras
3.2.3
The Kears Functional API
3.2.4
Models as Directed acyclic graphs of layers
3.3
Monitoring deep learning models
3.3.1
Using callbacks
3.3.2
TensorBoard
3.4
Batch normalization
3.5
Overfitting handling
3.5.1
Reducing the network’s size
3.5.2
Adding weight regularization
3.5.3
Adding dropout
3.5.4
Data augmentation
3.6
Hyperparameters optimization
4
Deep learning for computer vision
4.1
Image classification with Keras
4.1.1
download and prepare the data
4.1.2
Build the model
4.1.3
Compile the model
4.1.4
Fit the model
4.1.5
Make predictions
4.1.6
Evaluate the model
4.1.7
Save the model
4.1.8
reload the model
4.2
Introduction to Convolution Neural Networks
4.2.1
Example
4.2.2
The convolution operation
4.2.3
The max-pooling operation
4.3
Architectures of CNN
4.4
Classifcation examples
4.4.1
Dataset: Dog VS Cats
4.4.2
Dataset: CIFAR10
5
Transfer learning
5.1
Introduction
5.1.1
What is Transfer Learning
5.1.2
What is TensforFlow Hub
5.2
Transfer learning with TensforFlow Hub
5.2.1
ImageNet classifier
5.2.2
Transfer learning
5.2.3
Run the classifier on a batch of images
5.2.4
Download a headless model
5.2.5
Attach a classification head
5.2.6
Train the model
5.2.7
Export the model
5.3
Transfer learning using a pretrained CONVNET
5.3.1
feature extraction
5.3.2
Fine-tuning
6
Visualizing what CONVNETS leran
References
Published with bookdown
Computer vision with R and keras
References