Hey everyone ! welcome to my blog. This blog covers detailed instructions to building a docker image to pretrain ELECTRA. I can hear you asking why do I need to build a docker image? Okay here it goes. The required version for running ELECTRA is tensorflow 1.15.5 (which is older…

This blog gives a detailed explanation of using DVC for version data and models and also dockerizing the application. If you are interested in learning how to do the version control using DVC you can read the blog here. In the previous blog Versioning data and models in ML projects…

Hi Everyone ! Welcome to my blog. In this blog, I am going to show a simple implementation of SimCSE: Simple Contrastive Learning of Sentence Embeddings for the unsupervised approach. In SimCSE, the authors have used a simple contrastive learning framework to generate state-of-the-art sentence embeddings. In the supervised approach…

Welcome to the second part on optimisers where we will be discussing momentum and Nesterov accelerated gradient. If you want a quick review of vanilla gradient descent algorithms and its variants, please read about it in part1. …

Recent progress in the Natural Language Processing domain, such as the transformer based models have given us several State-of-the-Art (SOTA) pretrained models. These large pretrained models can then be fine-tuned on the custom dataset for specific tasks. …

We will be looking at how DVC can be used to version our data and models in this blog in detail. The code for this blog is available here. For details regarding the model training for Named Entity Recognition (NER) tagging of sentences (CoNLL-2003 dataset ) using Tensorflow2.2.0, …

This blog details the steps for fine-tuning the BERT pretrained model for Named Entity Recognition (NER) tagging of sentences (CoNLL-2003 dataset ). If you are new to NER, i recommend you to go through this NER for CoNLL dataset with Tensorflow 2.2.0 blog first. Let us see how we can…

This blog details the steps for Named Entity Recognition (NER) tagging of sentences (CoNLL-2003 dataset ) using Tensorflow2.2.0 CoNLL-2003 dataset includes 1,393 English and 909 German news articles. We will be looking at the English data. The CoNLL-2003 data files contain four columns separated by a single space. Each word…

INSTALLING PYDRIVE On your remote machine where you want to download your files from Google drive, install PyDrive with pip command $ pip3 install PyDrive 2. OAuth AUTHENTICATION : Downloading client_secrets.json file. Drive API requires OAuth2.0 for authentication. Go to APIs Console and create your project. When you click on…