Projects using TensorFlow in Punjab

Projects using TensorFlow in PunjabProjects using TensorFlow in Punjab

Cover image for Trained a DreamBooth LoRA model

Trained a DreamBooth LoRA model to generate high-quality, personalized image outputs with consistent subject identity across different prompts and styles. The project involved dataset preparation, image captioning, and fine-tuning diffusion models using LoRA for efficient training and deployment. The solution enables fast generation of customized visuals while preserving subject consistency, style control, and high fidelity, suitable for creative, branding, and content generation use cases.

1

170

Cover image for O-A: Predictive Demand Dashboard

O-A: Predictive Demand Dashboard

0

11

Cover image for Shopcast - Personalized Recommendation Engine

Shopcast - Personalized Recommendation Engine for Shopify: designed a collaborative-filtering recommendation engine that personalizes product suggestions from shopper behavior (views, add-to-cart, purchases, engagement). Built with Django, TensorFlow, BigQuery and Redis caching for real-time storefront delivery. Published live on the Shopify App Store, lifting merchants' conversion rate and average order value.

0

10

Cover image for Earthquake prediction using LSTM

Earthquake prediction using LSTM

0

86

Cover image for ARA Signature Forgery Detection Web Application

ARA Signature Forgery Detection Web Application

0

44

Cover image for Dill Plant Disease Detection

Dill Plant Disease Detection

0

2

Cover image for Real Time SMS Spam Detection/Classification

Real Time SMS Spam Detection/Classification System

1

152

Cover image for Yeah, this one is a

Yeah, this one is a bit out there, but it was super fun. I built an AI that actually understands memes. It looks at the image and the text together using PyTorch, TensorFlow, and GPT 4. I used early and late fusion techniques to combine vision and language, plus Tesseract OCR to grab text from the image itself. The model hits over 95% accuracy on meme classification. It’s not your typical corporate project, but it shows how multimodal AI can understand humor, culture, and context. Definitely one of my favorite experiments. For more details, please visit: https://arham-nexus.vercel.app/work/memechecker

1

0

89

Cover image for AI Neural Network for Cats vs. Dogs Classification

AI Neural Network for Cats vs. Dogs Classification

1

10

Cover image for Deep Learning

0

8

Cover image for Pixel Character Sprites Generation

Pixel Character Sprites Generation

0

34

Cover image for MCTS-UCB Transformer Ensemble

MCTS-UCB Transformer Ensemble

0

9

Cover image for At Formulatrix, I led the

At Formulatrix, I led the development of a computer vision pipeline for RockMaker, a biotech product used in crystallization experiments. I built and deployed object detection and image classification models that improved scoring accuracy by 13%, reducing manual effort for scientists and increasing customer satisfaction. The solution was productionized with Docker and CI/CD pipelines, ensuring scalability and reliability across client sites.

0

330

Cover image for An end-to-end OCR pipeline combining

An end-to-end OCR pipeline combining YOLOv11s for real-time document region detection with EasyOCR for multilingual text extraction from scanned documents and images.

0

25

Cover image for Identification of fabric defects using Deep Learning

Identification of fabric defects using Deep Learning

0

26

Cover image for Facial Emotion Recognition AI
Designed and

Facial Emotion Recognition AI Designed and trained a custom CNN to classify human facial expressions into 7 categories using the FER-2013 dataset. Built an interactive Streamlit dashboard for real-time emotion prediction from images. Implemented class weighting to address dataset imbalance and improved model performance. Custom CNN Architecture Real-time Dashboard 84% Accuracy Technologies Python TensorFlow Keras CNN Streamlit

1

168