PyTorch introductory tutorial notebooks and codes. Useful for deep learning practitioners.
- PyTorch_Tensors_Intro
Introduction to PyTorch Tensors (dimensions, broadcasting, basic linear algebra and memory operations)
- PyTorch_Autograd_Intro
Introduction to automatic differentiation engine of PyTorch. Autograd automatically tracks the arithmetic operations and computes the derivates
- PyTorch_Polynomial
Toy polynomial dataset and model. An example demostration of learning loops in PyTorch.
- PyTorch_Models_Intro
Basic linear models (implemented as nn.Module).
- PyTorch_Iris_Classification
Linear classifier model that works on tabular datasets. Toy dataset Iris is used to train the model.
- PyTorch_FlatNet_MNIST
Image classifier model that uses flatted pixels of images. Implemented with nn.Linear, does not use any convolution layers. Trained with built-in MNIST dataset.
- PyTorch_Conv_MNIST
Image classifier model that uses convolution layers. Trained with built-in MNIST dataset.
- PyTorch_Conv_ImageFolder
Image classifier model that uses convolution layers. Trained with an external dataset using ImageFolder Dataset class of PyTorch.
- PredictionUI_MNIST_FlatNet, PredictionUI_MNIST_Conv, PredictionUI_ImageFolder
Gradio UI applications for the models above. (NOTE: you need to train the models first.)
- NLP_Intro
Basics of Natural Language Processing (NLP) with text preprocessing. Vectoral representations of text. Introduction to spaCy with Named entity recognition (NER) and keyword Extraction using Part of Speech (POS) (using NNP tags).
- PyTorch_Word2Vec_CBOW
Continuous Bag of Words (CBOW) implementation of Word2Vec in PyTorch.