The most practical use of word embeddings (word2vec, glove, etc) you will ever see. — King — Man + Woman = Queen You might have seen the traditional word2vec or Glove word embeddings examples that show King -Man+Woman = Queen. Here Queen will be returned from the word embedding algorithm given the words King, Man, and Woman. Today we will see how we can use this structure to solve a real-world problem.

A practitioner guide for when to use different text decoding strategies. — If you have worked with text generation models you would have encountered several decoding strategies like Greedy Search, Beam Search, Diverse beam search, Topk Sampling, TopP sampling, TopKP sampling. Here is a guide on when to use each of these: TLDR: Greedy, Beam Search: Translation, Summarization Diverse Beam Search: Paraphrasing, Image…

Exploring OpenAI Dall-E 2’s edit image mode for memes — In this blog, we will see an interesting application by OpenAI’s Dall-E 2 where you can use existing memes and create more meme template variations by using the edit image feature of Dall-E 2. So why do we need to create multiple variations of memes in the first place? We are building an AI SaaS app Supermeme.ai — An AI-powered meme generator to generate original…

Affordable AI/ML inference — ML training and inference is a hard problem. There might be several reasons why you want to offload your inference and training to other platforms than self-managing. Here is a list of platforms that you can explore if you are looking for economical AI inference - banana.dev Ship ML to…

How we built in public and became the #1 product of the day! — We launched our app Supermeme.ai — AI-powered Meme Generator on Product Hunt and became the #1 product of the day for May 5th, 2022. Being bootstrapped, a 3-month-old app, the team building this part-time alongside their full-time jobs, and considering it was my first every Product Hunt launch, coming #1…

Use NLP to generate wrong answers for MCQs in Edtech — If you are working at the crossroads of NLP and Edtech, you will sooner or later encounter the problem of generating distractors (wrong answer choices) for a given question and answer, automatically using NLP. What are distractors? Distractors are the wrong answers to a multiple-choice question. For example, if a given multiple choice…

Paraphrase to retain particular keywords for SEO and copywriting — Huggingface recently introduced guiding text generation with constrained beam search in the Transformers library. You can give guidance about which words need to be included in the decoded output text with constrained beam search. This has a few interesting use-cases in copywriting and SEO. Usecase 1: SEO (Search Engine Optimization)

Reduce GPT-3 prompt tokens and save dollars — GPT-3's highest and the most accurate model Davinci costs 6 cents for every 1000 tokens. So it isn’t really inexpensive to operate at scale in a production app. So beyond designing prompts, it is essential to even master the craft of smart prompting, that is to reduce the number of…