Pytorch Data Augmentation Tutorial

We work directly with hundreds of publishers to connect you with the right resources to fit your needs. In this blog post, we discuss how to train a U-net style deep learning classifier, using Pytorch, for segmenting epithelium versus stroma regions. Your training set may have certain images of particular form , example - in cat images , cat may appear centrally in the image. Before you start the tutorial below, you should. Now that we've learned about the basic feed forward, fully connected, neural network, it's time to cover a new one: the convolutional neural network, often referred to as a convnet or cnn. 04, Amazon Linux 2, and Amazon Linux. Earlier this year, we shared a vision for making AI development faster and more interoperable. Welcome to PyTorch Tutorials¶. PyTorch Tutorial: PyTorch CIFAR10 - Load CIFAR10 Dataset (torchvision. Be sure to check for the types to avoid Type compatibility errors. jtiger958/pytorch-computer-vision-tutorial. The library respects the semantics of torch. Automatic Differentiation is a building block of not only PyTorch, but every DL library out there. PyTorch includes a package called torchvision which is used to load and prepare the dataset. Learn how to run your PyTorch training scripts at enterprise scale using Azure Machine Learning's PyTorch estimator class. 0 version, click on it. 25 in the transformed HSV image. It’s very easy to use GPUs with PyTorch. See ROCm install for supported operating systems and general information on the ROCm software stack. Explain Data Augmentation? What are its uses? Ans: Data augmentation adds value to base data by adding information derived from internal and external sources within an original. Although PyTorch Geometric already contains a lot of useful datasets, you may wish to create your own dataset with self-recorded or non-publicly available data. This feature is not available right now. Also drop us a comment on the tutorials that you'd love to read, I will try to have that up ASAP. This is a quick guide to run PyTorch with ROCm support inside a provided docker image. Optional: Data Parallelism¶. I^2C I^2C (pronounced 'I-squared-C') is also useful for communicating, but I have never used it. 04 with Volta GPU support [ Tutorials ] (3). PyTorch tutorials. Azure Machine Learning offers web interfaces & SDKs so you can quickly train and deploy your machine learning models and pipelines at scale. Segmentation in 3D using U-Nets with Delira - A very short introduction¶. The idea of doing image analysis is to. A place to discuss PyTorch code, issues, install, research. So now that you know the basics of what Pytorch is, let's apply it using a basic neural network example. PyTorch is a machine learning library which is based on Torch library. 2 can be used in the Azure platform. The primary reason for this is that the other transformations are applied on the input which is a PIL image, however, this must be converted to a PyTorch tensor before applying normalization. There are some issues about it on their github page. PyTorch is a relatively new neural network library which offers a nice tensor library, automatic differentiation for gradient descent, strong and easy gpu support, dynamic neural networks, and is easy to debug. In Tutorials. We'll then use Caffe2's mobile exporter to execute it on mobile. We need to pass data, calculate the loss function and modify network weights accordingly. This tutorial will walk you through the key ideas of deep learning programming using Pytorch. This tutorial describes the application of Singular Value Decomposition or SVD to the analysis of sparse data for the purposes of producing recommendations, clustering, and visualization on the Kinetica platform. Let’s see how it works! Here we:. nn as nn from torch. I tried convoluted NNets to model MNIST data recently, and it helped me in that case. You can do it from the terminal with:. Data augmentation: generate more training samples from original image. This tutorial is targeted at researchers who have some experience with spectral data collection and processing. Also drop us a comment on the tutorials that you'd love to read, I will try to have that up ASAP. cifar10) from Torchvision and split into train and test data sets. Hands-on tour to deep learning with PyTorch. What are Caffe2 and ONNX? Creating a Super-Resolution Model in PyTorch. PyTorch is basically exploited NumPy with the ability to make use of the Graphic card. The library is still under active development and supports fast data augmentation for all major ML development libraries out there — PyTorch, Tensorflow, MXNet. Here you can apply data augmentation to either, or both, your inputs and targets. Let’s create three transforms: Rescale: to scale the image. Fig 1: A typical data augmentation pipeline. We will use two different data generators for train and validation folders. Hybrid Front-End. In this blog post, we discuss how to train a U-net style deep learning classifier, using Pytorch, for segmenting epithelium versus stroma regions. This package implements an approach for missing view and missing data imputation via generative adversarial networks (GANs), which we name as VIGAN. Jeremy Howard’s wonderful tutorial on the PyTorch website is a good starting point. PyTorch provides many tools to make data loading easy and hopefully, to make your code more readable. Please visit this page for more information. The tutorial introduces Lasagne, a new library for building neural networks with Python and Theano. In this tutorial, we will see how to load and preprocess/augment data from a non trivial dataset…. The major difference from Tensorflow is that PyTorch methodology is considered "define-by-run" while Tensorflow is considered "defined-and-run", so on PyTorch you can for instance change your model on run-time, debug easily with any python debugger, while tensorflow has always a graph definition/build. 02:30pm Data Augmentation/Hallucination-based Techniques. Knowing that I was going to write a tutorial on. This article and video tutorial will help you get up to speed with generating synthetic training images in Unity. In the tutorial, most of the models were implemented with less than 30 lines of code. In this post, we describe how to do image classification in PyTorch. This is a two part article. For training… the difference is massive. GPUs offer faster processing for many complex data and machine learning tasks. This CentOS 7. PyTorch Tutorial: PyTorch CIFAR10 - Load CIFAR10 Dataset (torchvision. I also show a ton of use cases for different transforms applied on Grayscale and Color images, along with Segmentation datasets where the same transform should be applied to both the input and target images. Getting Started with Distributed Data Parallel¶. py & data_augmentation. Deep Learning with PyTorch: A 60 Minute Blitz # Data augmentation and normalization for training # Just. The program is part of the computer vision research and development for the Peter Moss Acute Myeloid/Lymphoblastic (AML/ALL) Leukemia AI Research Project. AWS Deep Learning AMIs now come with PyTorch 1. Image Classification. At the launch of Houdini 17. 9 release of Snorkel. PyTorch is an open source machine learning library based on the Torch library, used for applications such as computer vision and natural language processing. I am using the following code to do data augmentation of MNIST: Browse other questions tagged pytorch mnist data-augmentation or ask your own (MNIST Tutorial) 5. I have searched around the internet for some guides on how to import a image based data-set into Pytorch for use in a CNN. PyTorch Geometric makes implementing Graph Neural Networks a breeze (see here for the accompanying tutorial). 5 / PilotPDG we showcased PDG's application in Machine Learning in the form of data synthesis, data augmentation, and hyper-parameter space search. Pytorch-C++ is a simple C++ 11 library which provides a Pytorch-like interface for building neural networks and inference (so far only forward pass is supported). (BTW, by Keras I mean no boilerplate, not overly-simplified). Pyro is a universal probabilistic programming language (PPL) written in Python and supported by PyTorch on the backend. In Tutorials. We also use the PointsItemList class to have the targets be of type ImagePoints (which will make sure the data augmentation is properly applied to them). The position listed below is not with Rapid Interviews but with ASAP (Allied Staff Augmentation Partners, Inc. DistributedDataParallel (DDP) implements data parallelism at the module level. Also try practice problems to test & improve your skill level. Contribute to pytorch/tutorials development by creating an account on GitHub. 雷锋网 AI 开发者按:近日,PyTorch 社区又添入了「新」工具,包括了更新后的 PyTorch 1. Data augmentation : Data augmentation is a process where you make changes to existing photos like adjusting the colors , flipping it horizontally or vertically , scaling , cropping and many more. Tutorial on MatConvNet Bolei Zhou Leverage the object labels in some training data. I’ve found that facebookresearch/visdom works pretty well. This is the twelfth tutorial in this series. 4 ML developers who are using TensorFlow. After completing this tutorial, you will know: Image data should probably be centered by subtracting the per-channel mean pixel values calculated on the training dataset. I have taken this section from PyTorch-Transformers' documentation. This is a hands on tutorial which is geared toward people who are new to PyTorch. Transfer Learning. PythonProgramming. GAN concept and theory. You will also receive a free Computer Vision Resource Guide. GitHub Gist: instantly share code, notes, and snippets. Lab 2: Train a CNN on CIFAR-10 Dataset ENGN8536, 2018 August 13, 2018 In this lab we will train a CNN with CIFAR-10 dataset using PyTorch deep learning framework. PyTorch is an open source machine learning library based on the Torch library, used for applications such as computer vision and natural language processing. data and coco_100img. Once author Ian Pointer helps you set up PyTorch on a cloud-based environment, you'll learn how use the framework to create neural architectures for performing operations on images, sound, text, and other types of data. Pytorch Implementation of Neural Processes¶ Here I have a very simple PyTorch implementation, that follows exactly the same lines as the first example in Kaspar's blog post. 将数据级封装成类Dataset. This example demonstrates how you can use kubeflow end-to-end to train and serve a distributed Pytorch model on a kubernetes cluster in GCP. Deep Learning with PyTorch: A 60 Minute Blitz # Data augmentation and normalization for training # Just. Read through our online tutorials on data analysis & interpretation. PyTorch Geometric is a geometric deep learning extension library for PyTorch. We are then going to compute PCA on these data points. The method of choice to train on similar but different examples to the training data is known as data augmentation (Simard et al. The second is slicing training data with slicing functions (SFs), in order to monitor and focus model attention on subsets of the training dataset where classification is more critical or difficult. In this chapter, we will focus more on torchvision. pyplot as plt from torch. (data augmentation) ToTensor : numpy 배열의 이미지를 torch 텐서로 바꾸어준다. Plan of Attack. In this session we will show a way to do that with Jupyter and Azure. You’ll also learn how to deploy PyTorch models into production via Caffe2 using ONNX. Please read this tutorial there. In this blog post, we discuss how to train a U-net style deep learning classifier, using Pytorch, for segmenting epithelium versus stroma regions. PyTorch vs Apache MXNet; Six 10-minute tutorials covering the core concepts of MXNet using the Gluon API. The program is part of the computer vision research and development for the Peter Moss Acute Myeloid/Lymphoblastic (AML/ALL) Leukemia AI Research Project. The Effectiveness of Data Augmentation in Image Classification using Deep Learning by Luis Perez, Jason Wang; About Me: Graduated with MS Data Science at USF and undergrad in Computer Science, I have 2 years of experience in building predictive and recommendation algorithms, and deriving business insights for finance and retail clients. See ROCm install for supported operating systems and general information on the ROCm software stack. There will be an equal balance of theory and hands-on PyTorch coding. PyTorch Tutorial is designed for both beginners and professionals. The idea of doing image analysis is to. The primary reason for this is that the other transformations are applied on the input which is a PIL image, however, this must be converted to a PyTorch tensor before applying normalization. Python has a variety of packaging systems, all of which have good and not-so-good points. CIFAR-10 dataset contains 50000 training images and 10000 testing images. But good data doesn't grow on trees, and that scarcity can impede the development of a model. I am a little bit confused about the data augmentation performed in PyTorch. Yesterday, at the PyTorch Developer Conference, Facebook announced the release of PyTorch 1. A machine learning craftsmanship blog. On the first day of this training, you will be provided with a link to a JupyterHub instance where the environment will be pre-made and ready to go!. In this Python tutorial on Python Variables and Python Data Types, we will learn about at Python variables and data types being used in Python. But maybe we can make some. 【PyTorch: Tutorial 初級: データ並列 (マルチ GPU)】 PyTorch のドキュメントが改訂されていますので、再翻訳しています。最新の PyTorch 0. PyTorch Tutorial for Beginner CSE446 Department of Computer Science & Engineering University of Washington February 2018. 最后,安利一下pytorch中文文档。 在各方小伙伴的努力和支持下,pytorch中文文档第一版终于上线啦! (鼓掌)文档还有很多小瑕疵,但是大体可以放心使用了~我们遵循快速迭代的原则,所以赶紧上线第一版来接受广大开源社区的意见和建议。. In this tutorial, I give an example of CUDA extension for data augmentation in deep learning. I found out data augmentation can be done in PyTorch by using torchvision. She has also been working on open AI systems and is one of the co-creators of ONNX, an open standard for deep learning models, and a leader in the Pytorch 1. In our previous PyTorch notebook, we learned about how to get started quickly with PyTorch 1. Using Tutorial Data from Google Drive in Colab — PyTorch Tutorials 1. These techniques are independent of the data type and we therefore use just one of the datasets to illustrate them. PyTorch provides a hybrid front-end that allows developers to iterate quickly on their models in the prototyping stage without sacrificing performance in the production stage. Weâ??ll cap it off by tackling a more complicated example that uses convolutional neural networks (CNNs) to improve accuracy. In a blog post this week, the company discussed how the latest version of the. data augmentation in pytorch. A universal and efficient framework for training well-performing light net; A utility library around PyTorch; Comprehensive Data Augmentation and Sampling for Pytorch. Just make sure your controller has some method of communicating data to you for easy and effective debugging/testing of your robot programs. com Abstract In this paper, we explore and compare multiple solutions to the problem of data augmentation in image classification. Deep Learning with Pytorch -CNN from Scratch with Data Augmentation – 2. I also show a ton of use cases for different transforms applied on Grayscale and Color images, along with Segmentation datasets where the same transform should be applied to both the input and target images. But i have. Automatic Differentiation is a building block of not only PyTorch, but every DL library out there. In this section, we will see how to build and train a simple neural network using Pytorch tensors and auto-grad. High-Level Training, Data Augmentation, and Utilities for Pytorch. In this blog I will offer a brief introduction to the gaussian mixture model and implement it in PyTorch. A PyTorch tensor is a specific data type used in PyTorch for all of the various data and weight operations within the network. Stay tuned! Have you used PyTorch to build an application or in any of your data science projects?. Please try again later. 4-based data science virtual machine (DSVM) contains popular tools for data science and development activities, including Microsoft R Open, Anaconda Python, Azure command line tools, and xgboost. datasets and its various types. I have been learning it for the past few weeks. As a simple example, a linear regression model with a single parameter has a much lower capacity than a linear regression model with multiple polynomial parameters. Welcome to part 5 of the TensorFlow Object Detection API tutorial series. A small bit of code in the dataset class was also needed to be changed to assert this tensor type on the pixel data as the current version of PyTorch didn't seem to apply the newly set default. My aim here is to Explain all the basics and practical advic. The Dataset API allows you to build an asynchronous, highly optimized data pipeline to prevent your GPU from data starvation. PyTorch Tutorial. Load CIFAR-10 dataset from torchvision. PyTorch provides. In this release we introduced many exciting new features and critical bug fixes, with the goal of providing users a better and cleaner interface. It is primarily used for applications such as natural language processing. 10 Minute Viewshare Tutorial One of the best parts of Viewshare is that it can actually enhance your data through a process we call augmentation. Author: Shen Li. GitHub Gist: instantly share code, notes, and snippets. Author: Sasank Chilamkurthy. But maybe we can make some. datasets and torch. Even though it is possible to build an entire neural network from scratch using only the PyTorch Tensor class, this is very tedious. Facebook today announced the release of deep learning framework PyTorch 1. It is primarily developed by Facebook's artificial intelligence research group. All the other code that we write is built around this- the exact specification of the model, how to fetch a batch of data and labels, computation of the loss and the details of the optimizer. Getting Started with Distributed Data Parallel¶. PyTorch provides many tools to make data loading easy and hopefully, to make your code more readable. Introducing Pytorch for fast. ai course the focus was to allow student so read and implement recent research papers, and pytorch made this easier due to its flexibility. 8% improvement in the mAP. 01:30pm Introduction (Rogerio Feris). You will use the Keras deep learning library to train your first neural network on a custom image dataset, and from there, you’ll implement your first Convolutional Neural Network (CNN) as well. Sometimes our models overfit, sometimes they overfit. Even though it is possible to build an entire neural network from scratch using only the PyTorch Tensor class, this is very tedious. PyTorch, Facebook's deep learning framework, is clear, easy to code and easy to debug, thus providing a straightforward and simple experience for developers. A small bit of code in the dataset class was also needed to be changed to assert this tensor type on the pixel data as the current version of PyTorch didn't seem to apply the newly set default. NVIDIA DALI: A GPU-accelerated data augmentation and image loading library for optimizing data pipelines of deep learning frameworks NVIDIA nvJPEG: A high-performance GPU-accelerated library for JPEG decoding Computer vision applications powered. The main process takes data from the pipe, makes batches. DataParallel module. It is primarily used for applications such as natural language processing. In our previous PyTorch notebook, we learned about how to get started quickly with PyTorch 1. In this part, we will implement a neural network to classify CIFAR-10 images. Be sure to check for the types to avoid Type compatibility errors. Train a simple deep CNN on the CIFAR10 small images dataset. End-to-End kubeflow tutorial using a Pytorch model in Google Cloud. There will be some overlap with last year’s DAS 2018 tutorial, and the tutorial will provide a self-contained introduction, but the focus will be on different topics, including the latest version of PyTorch, large scale processing, distillation, semi-supervised training, and distributed training. As an example, we will build an image. Let us now look at the ways of selecting data variables. PyTorch includes a package called torchvision which is used to load and prepare the dataset. 0 early this year with integrations for Google Cloud, AWS, and Azure Machine Learning. You will also receive a free Computer Vision Resource Guide. The very first thing we have to consider is our data. If you’ve never used PyTorch or any machine learning framework before, take a look at this tutorial, which goes over the basic operations and some simple models. Data augmentation and preprocessing is an important part of the whole work-flow. An extensive collection of articles about NCBI databases and software. /data', train=False, download=True. The major difference from Tensorflow is that PyTorch methodology is considered "define-by-run" while Tensorflow is considered "defined-and-run", so on PyTorch you can for instance change your model on run-time, debug easily with any python debugger, while tensorflow has always a graph definition/build. First download the data from here and unzip it in the working directory on the Jetson. Pytorch tutorials for Neural Style transfer. Data Parallelism in PyTorch for modules and losses - parallel. Behind her. PyTorch is a promising python library for deep learning. The curve in linear regression follows a Machine Learning Linear Regression Pytorch Primer Tutorial. Learning or getting started with PyTorch is as easy as creating your Azure account and cloning the tutorial notebooks into your own library. Knowing that I was going to write a tutorial on. You can choose a variety of machine types to host your Data Science Virtual Machine, including those with GPUs. Let’s talk about data augmentation next. PyTorch is one of many packages for deep learning. 3, which has been used for exporting models through ONNX. You could also, as said in the tutorial itself, also look at some other data augmentation operations that you can perform on the traffic sign images. This post is intended to be useful for anyone considering starting a new project or making the switch from one deep learning framework to another. Earlier this year, we shared a vision for making AI development faster and more interoperable. To get a better understanding of RNNs, we will build it from scratch using Pytorch tensor package and autograd library. deb based system. Be sure to check for the types to avoid Type compatibility errors. As a Python-first framework, PyTorch enables you to get started quickly, with minimal learning, using your favorite Python libraries. dev20180918 documentation について解説する。. Data Augmentation helps the model to classify images properly irrespective of the perspective from which it is displayed. But that doesn't seem like happening in PyTorch. Data is one of the core assets for an enterprise, making data management essential. Our Tutorial provides all the basic and advanced concepts of Deep learning, such as deep neural network and image processing. Select preferences and run the command to install PyTorch locally, or get started quickly with one of the supported cloud platforms. I would like to provide a short tutorial showing how to both generate data for a toy problem and use that data to train an image segmentation neural network. we all have been there. - So remember, what we're trying to achieve here…is we're going to look at our training data set…and try and manipulate as many of the images as possible…to alter them. Keras, TensorFlow and PyTorch are among the top three frameworks that are preferred by Data Scientists as well as beginners in the field of Deep Learning. In this tutorial, you’ll get an introduction to deep learning using the PyTorch framework, and by its conclusion, you’ll be comfortable applying it to your deep learning models. You could also, as said in the tutorial itself, also look at some other data augmentation operations that you can perform on the traffic sign images. NCBI Handbook. Using pytorch's torchvision. GitHub Gist: instantly share code, notes, and snippets. PyTorch Tutorials 0. On the first day of this training, you will be provided with a link to a JupyterHub instance where the environment will be pre-made and ready to go!. Facebook launched PyTorch 1. This efficient input pipeline is important in practice to achieve a high utilization of GPUs, esp. DistributedDataParallel (DDP) implements data parallelism at the module level. The main goal of word2vec is to build a word embedding, i. To do this, we need the Images, matching TFRecords for the training and testing data, and then we need to setup the. Pytorch is a dynamic instead of static deep learning library and Jeremy Writes that nearly all of the top 10 Kaggle competition winners now have been using Pytorch. Author: Justus Schock, Alexander Moriz. Pytorch tutorials for Neural Style transfer. PyTorch vs Apache MXNet¶. • Hand-designed data set augmentation can dramatically improve performance • When comparing ML algorithms A and B, same data set augmentation should be used for both – If A performs poorly with no dataset augmentation and B performs well with synthetic transformations of the input, reason may be the data set rather than algorithm. Also try practice problems to test & improve your skill level. According to a survey, there are 1,616 ML developers and data scientists who are using PyTorch and 3. SMOTE uses a k-Nearest. Setup network to train. This lesson showcases the data synthesis and augmentation aspects of PDG to further knowledge in this area. pytorch PyTorch 101, Part 2: Building Your First Neural Network. In the previous blog we discussed about PyTorch, it's strengths and why should you learn it. A lot of effort in solving any machine learning problem goes in to preparing the data. PyTorch vs Apache MXNet; Image Tutorials. This package implements an approach for missing view and missing data imputation via generative adversarial networks (GANs), which we name as VIGAN. I also show a ton of use cases for different transforms applied on Grayscale and Color images, along with Segmentation datasets where the same transform should be applied to both the input and target images. It can also be used for shallow learning, for optimization tasks unrelated to deep learning, and for general linear algebra calculations with or without CUDA. Creating Your Own Datasets¶. when I/O operations and data augmentation is a bit heavy. By randomly cropping images to be 24x24 you can still keep the main object (the neural network won’t be confused to see abnormal images) and save computing. Inside this Keras tutorial, you will discover how easy it is to get started with deep learning and Python. Skip to content. All gists Back to GitHub. Azure Machine Learning offers web interfaces & SDKs so you can quickly train and deploy your machine learning models and pipelines at scale. The Symbol API in Apache MXNet is an interface for symbolic programming. Last week, Hugging Face, a startup specializing in natural language processing, released a landmark update to their popular Transformers library, offering unprecedented compatibility between two major deep learning frameworks, PyTorch and TensorFlow 2. Models from pytorch/vision are supported and can be easily converted. abstract_callback import AbstractCallback if 'TORCH' in get_backends (): from torch. The project does not actually include the car simulator, and the actual example it does come with is not suitable for generating training data. The course will teach you how to develop Deep Learning models using Pytorch while providing the necessary deep-learning background. To use cuda (and cudnn), make sure to set paths in your. PyTorch includes a package called torchvision which is used to load and prepare the dataset. device("cuda: 0" if torch. 5 / PilotPDG we showcased PDG's application in Machine Learning in the form of data synthesis, data augmentation, and hyper-parameter space search. In this tutorial, we will learn how to use multiple GPUs using DataParallel. fastai provides a complete image transformation library written from scratch in PyTorch. 3, which has been used for exporting models through ONNX. Well, we already have a set of images. 3% for a while. Attributes. 0 was released this week one of the major things it introduced was a new C++ Front-End , the ability to build models using C++, with a similar API to PyTorch. At training time we will perform the # following preprocessing on our images: # (1) Resize the image so its smaller side is 256 pixels long # (2) Take a random 224 x 224 crop to the scaled image # (3) Horizontally flip the image with probability 1/2. cifar10) from Torchvision and split into train and test data sets. This repository provides tutorial code for deep learning researchers to learn PyTorch. According to the paper, the use of data augmentation leads to a 8. A guide to data augmentation. Data augmentation: generate more training samples from original image. …So we go to images, sample-train,…and we see in our images. /data', train=False, download=True. I have designed this TensorFlow tutorial for professionals and enthusiasts who are interested in applying Deep Learning Algorithm using TensorFlow to solve various problems. Also drop us a comment on the tutorials that you'd love to read, I will try to have that up ASAP. Many of the concepts (such as the computation graph abstraction and autograd) are not unique to Pytorch and are relevant to any deep learning tool kit out there. In this post I'm going to present library usage and how you can build a model using our favorite programming language. PyTorch is one of many packages for deep learning. See more of Dataiku on Facebook. For examples and more information about using PyTorch in distributed training, see the tutorial Train and register PyTorch models at scale with Azure Machine Learning. This document proposed two major changes to the current ARPANET host access protocol. PyTorch provides a hybrid front-end that allows developers to iterate quickly on their models in the prototyping stage without sacrificing performance in the production stage. Models from pytorch/vision are supported and can be easily converted. x 가 Variable일 때, x. The main goal of word2vec is to build a word embedding, i. The AWS Deep Learning AMIs are now available on Ubuntu 18. In this tutorial, you will learn how the Keras. Smart approaches to programmatic data augmentation can increase the size of your training set 10-fold or more. With data augmentation we can flip/shift/crop images to feed different forms of single image to the Network to learn. In the next few articles, I will apply PyTorch for audio analysis, and we will attempt to build Deep Learning models for Speech Processing.