Frameworks for neural network modeling
Multiple frameworks have been used for neural network modeling:
- PyTorch (https://pytorch.org/)
- TensorFlow (https://www.tensorflow.org/learn)
- Keras (https://keras.io/)
- Caffe (https://caffe.berkeleyvision.org/)
- MXNet (https://mxnet.apache.org/versions/1.9.1/)
In this book, we will focus on PyTorch in practicing deep learning, but the concepts we’ll introduce are independent of the framework you use in your projects.
PyTorch for deep learning modeling
PyTorch is an open source deep learning framework, based on the Torch library, developed by Meta AI. You can easily integrate PyTorch with Python’s scientific computing libraries in your deep learning projects. Here, we will practice using PyTorch by looking at a simple example of building an FCNN model using the MNIST digit dataset. It is a commonly used example and the objective is solely to understand how to train and test a deep learning model using...
