PyTorch Fundamentals

1. Introduction to PyTorch:

   • Learn how to build machine learning models with PyTorch.

2. Introduction to Computer Vision with PyTorch:

   • Get an introduction to Computer Vision using one of the most popular deep learning frameworks, PyTorch! 

   • We'll use image classification tasks to learn about convolutional neural networks, and then see how pre-trained networks and transfer learning can improve our models and solve real-world problems.

3. Introduction to Natural Language Processing with PyTorch:

   • Explore different neural network architectures for dealing with natural language texts. 

   • In the recent years, Natural Language Processing (NLP) has experiences fast growth as a field, primarily because performance of the language models depend on their overall ability to "understand" text, and that can be trained in unsupervised manner on large text corpora. Thus, pre-trained text models such as BERT simplified many NLP tasks, and dramatically improved the performance.

4. Introduction to Audio Classification with PyTorch:

   • Learning how to do audio classification with PyTorch.

   • There are multiple ways to build an audio classification model. You can use the waveform, tag sections of a wave file, or even use computer vision on the spectrogram image. In this tutorial we will first break down how to understand audio data, from analog to digital representations, then we will build the model using computer vision on the spectrogram images. Thats right, you can turn audio into an image representation and then do computer vision to classify the word spoken!

Tips to Improve Pytorch: 

1. Maximize DataLoader: num_workers, batch_size.

2. 16-bit Precision: auto_cast, GradScaler.

3. Accumulated Gradients.

4. Retained Graphs.