Deployment, observability, evaluation, infrastructure, and operations.
Objective: Show a basic architecture and implementation of a CV (computer vision) CNN model to extract product attributes by using object detection, by training custom product image data on large pre-trained CNN…
Testing #GPT4o, good but mac app has inference latency issues. #multimodal is the future. #LLMs are commoditized. Fine tuned #industry specific models with #device level tuning inference + cloud generalization seems more…
Sharing personal github repo for code, theory, math’s and projects for training on all aspects of #SI, #ML, Agentic SI, #datascience research for solving real world problems using classical #supervised and #DeepLearning…
I am building #multimodal #opensource model for #healthcare or any industry is challenging but fun #technically. So much to learn apply. Data is issue but #synthetic distribution generation with human validation works.…
Predict 3 year Forward Asset Prices & Direction and Portfolio Return & Risk using Historic Macro data & Central Bank Forecasts This algorithm was published earlier. Sharing it again as it is…
This article will discuss the basic theory and relevant code examples for different clustering algorithms. Code for different algorithms and data used are given in github. Note – Make sure to point…
This article will discuss the basic theory and relevant code examples for different classification algorithms. It will also show how to test and compare multiple classification algorithms at once on the same…
This article will discuss the basic theory and relevant code examples for different regression algorithms. It will also show how to test and compare multiple regression algorithms at once on the same…
Common NLP tasks (Sentiment, Keyword, Topic, Summary, etc) using main libraries (HuggingFace Transformers, Spacy, Pytorch, Tensorflow, Scikit, etc) Idea is to show how to perform different NLP (Natural Language Processing) tasks using…
Using Tensorflow and keras for building CNN model with CIFAR-10 dataset for image classification Built different CNN models using different techniques and layersAchieved an accuracy of more than 85% between train and…
Computer Vision model for livestream frame people counter. Using OpenCV, YOLO5 from ultralytics, pytorch YOLO5 used the COCO image dataset for training its model Logic: The OpenCV library is used to get…
Interesting article on data engineers vs data scientists from https://gradientflow.com/ is given below. One of the main issues for building ML/SI/DL models is availability of good, clean, relevant and large datasets. Very…
