Switch branches/tags
Nothing to show
Find file History
Permalink
..
Failed to load latest commit information.
numerai conv1d working Sep 12, 2017
01 PyTorch GPU support test.ipynb CUDNN 6 Sep 30, 2017
02 PyTorch basic Tensor operations.ipynb PyTorch upgrade Nov 8, 2017
03 PyTorch CPU to GPU copy.ipynb conv1d working Sep 12, 2017
05 PyTorch Automatic differentiation.ipynb Torch Nov 8, 2017
07 PyTorch two layer nn.ipynb conv1d working Sep 12, 2017
08 PyTorch Image Classification using Pre-trained squeezenet1_1.ipynb PyTorch upgrade Nov 8, 2017
09 PyTorch Kaggle Image Data-set loading with CNN.ipynb PyTorch upgrade Nov 8, 2017
13 PyTorch Logistic Regression .ipynb Torch Nov 8, 2017
18-PyTorch-NUMER.AI-Binary-Classification-BCELoss-0.691839667509 .ipynb Torch Nov 8, 2017
18-PyTorch-NUMER.AI-Binary-Classification-BCELoss-0.69218508057.ipynb CNN conv1d for Numerai Sep 16, 2017
18-PyTorch-NUMER.AI-Binary-Classification-BCELoss.ipynb Torch Nov 8, 2017
21-PyTorch-CIFAR-10-Custom-data-loader-from-scratch.ipynb conv1d working Sep 12, 2017
31-PyTorch-using-CONV1D-on-one-dimensional-data.ipynb conv1d working Sep 12, 2017
500_-_The_request_timed_out_.png Add files via upload Nov 28, 2017
55-PyTorch-using-CONV1D-on-one-dimensional-data-CNN.ipynb Torch Nov 8, 2017
CIFAR-CLEAN.ipynb conv1d working Sep 12, 2017
IceResNet_2017-11-27_02-33-15.png Add files via upload Nov 28, 2017
Installation and troubleshooting guide-.md CNN conv1d for Numerai Sep 16, 2017
README.md Update README.md Sep 18, 2017
build_torch.sh CUDNN 6 Sep 30, 2017
data-03-diabetes.csv conv1d working Sep 12, 2017
data_util.py conv1d working Sep 12, 2017
day.png conv1d working Sep 12, 2017
nn-architecture.png Add files via upload Sep 18, 2017
pure_mnist.ipynb conv1d working Sep 12, 2017
pytorch-logo-flat.png conv1d working Sep 12, 2017

README.md

pytorch


Deep Learning Bootcamp November 2017, GPU Computing for Data Scientists: PyTorch

Web: https://www.meetup.com/Tel-Aviv-Deep-Learning-Bootcamp/events/241762893/

https://www.meetup.com/Tel-Aviv-Deep-Learning-Bootcamp/events/242418339/

Notebooks: On GitHub

pytorch

PyTorch is an optimized tensor library for Deep Learning, and is a recent newcomer to the growing list of GPU programming frameworks available in Python. Like other frameworks it offers efficient tensor representations and is agnostic to the underlying hardware. However, unlike other frameworks it allows you to create “define-by-run” neural networks resulting in dynamic computation graphs, where every single iteration can be different—opening up a whole new world of possibilities. Central to all neural networks in PyTorch is the Autograd package, which performs Algorithmic Differentiation on the defined model and generates the required gradients at each iteration.

Keywords: GPU Processing, Algorithmic Differentiation, Deep Learning, Linear algebra.


pytorch


Jupyter Notebooks

This repo contains the PyTorch implementations of various Deep Learning Algorithms.
Jupyter Notebooks are a very effective tool for interactive data exploration and visualisation.

List of Tutorials

Title Description
[Binary Classification with MLP]https://github.com/QuantScientist/Deep-Learning-Boot-Camp/blob/master/day%2002%20PyTORCH%20and%20PyCUDA/PyTorch/18-PyTorch-NUMER.AI-Binary-Classification-BCELoss.ipynb) NUMER.AI Deep Learning Binary Classification using BCELoss.
[[Binary Classification with CNN]https://github.com/QuantScientist/Deep-Learning-Boot-Camp/blob/master/day%2002%20PyTORCH%20and%20PyCUDA/PyTorch/55-PyTorch-using-CONV1D-on-one-dimensional-data-CNN.ipynb) NUMER.AI Deep Learning Binary Classification using CNN.

The HTML slides were created using:

%%bash jupyter nbconvert \ --to=slides \ --reveal-prefix=https://cdnjs.cloudflare.com/ajax/libs/reveal.js/3.2.0/ \ --output=py05.html \ './05 PyTorch Automatic differentiation.ipynb'

Project structure

The project consists of the following folders and files:

  • data/: contains Data sets used in the Jupyter Notebooks,
  • notebook/: collection of PyTorch Jupyter Notebooks for data exploration and results visualisation;

Dependencies

IDE

This project has been realised with PyCharm by JetBrains

GPU selection

Let's say your machine has N GPUs. You can choose to use any of these, by specifying the index n = 0, ..., N-1. Therefore, type CUDA_VISIBLE_DEVICES=n just before python ... in the following sections.

Workshop Agenda:

Module 1 Getting Started

  • What is Pytorch

  • Install and Run Pytorch

  • Allocating CPU Tensors using PyTorch

  • Allocating GPU Tensors using PyTorch

Module 2 Basic Pytorch Operations

  • Tensor Operation

  • Numpy Bridge

  • Variable

  • Gradients and Autograd

Module 3 Data Pre-processing

  • Install and Run Torchvision

  • Datasets

  • Data Transformation

Module 4 Linear/Logistic Regression with Pytorch

  • Loss Function

  • Optimizer

  • Training

Module 5 Neural Network (NN) with Pytorch

  • What is Neural Network

  • Activation Functions

  • Deep Neural Network with Pytorch

Module 7 Convolutional Neural Network (CNN) with Pytorch

  • What is CNN?

  • CNN Architecture

  • Convolution

  • Pooling and Stride

  • Dropout

Author

Shlomo Kashani/ @QuantScientist

A very comprehensice list of PyTorch links: