Steps to create, test and push a docker image Download and install Docker Desktop. If on Linux, download Docker Engine - Community. Create a Docker file with the name Dockerfile or download o...

LAMB paper Previous LR scaling with batch size Simple large batch training Training with extremely large batch was difficult. The researchers needed to carefully tune training hyper-parameter...

This document analyses the memory usage of Bert Base and Bert Large for different sequences. Additionally, the document provides memory usage without grad and finds that gradients consume most of t...

What are Transformers? Transformers are new state-of-the-art deep learning architectures used for Natural Language Processing(NLP). What is NLP? NLP comprises of AI tasks related to text process...

Contingency table Contingency table and chi squared distributions are used to determing if two categorical varibles are independent or not. For example: The following table shows a contingency tab...

Original Documentation Video link: https://channel9.msdn.com/Events/Connect/Microsoft-Connect–2018/D240/ Start by creating a free subscription in azure portal : https://portal.azure.com Create wo...

This is a GRU based RNN classifier to predict the read probability of a user from his/her email data. import torch import pandas as pd import numpy as np from torch.utils.data import Dataset,DataL...

#! pip install pytorch-pretrained-bert Using BERT BERT Paper Tokenizer See how BERT tokenizer works Tutorial source : Huggingface BERT repo import torch from pytorch_pretrained_bert import Bert...

A recurrent neural network (RNN) is a class of artificial neural network where connections between units form a directed cycle. This is a complete example of an RNN multiclass classifier in pytorc...

Word Embeddings Word embeddings are lower dimentional dense representation of words. Generally, embeddings are generated using semi supervised learning where the embeddings are learned based on an...