Reimplemented ”Sequence to Sequence Learning with Neural Networks” paper using TensorFlow on English-French WMT14\ndataset with 40,000 sentence pairs
\nDesigned 4-layer LSTM encoder-decoder architecture using Bahdanau attention mechanism to handle variable-length se-\nquences up to 50 tokens and 512-dimensional hidden state representations
\nOptimized training pipeline implementing gradient clipping, teacher forcing with decay schedule, and adaptive learning\nrate to reduce the convergence time by 30% and prevent gradient explosion
"}},{"node":{"frontmatter":{"date":"2025-11-01","title":"LifeLink - Organ Transplant Management System","tech":["MySQL","Database Design","SQL","Python","Algorithms"],"github":"https://github.com/sej07/LifeLink","external":""},"html":"Comprehensive database-driven organ transplant management system handling the complete donation lifecycle. Designed 18 normalized (3NF) tables with automated matching algorithms using weighted scoring (Blood: 30%, HLA: 25%, Wait Time: 20%). Implemented 6 automated triggers, 4 stored procedures, and 4 custom functions for real-time priority-based waitlist management and organ allocation with viability tracking.
"}},{"node":{"frontmatter":{"date":"2025-10-01","title":"CIFAR-10 Image Classification","tech":["PyTorch","CNNs","Python","Data Augmentation"],"github":"https://github.com/sej07/CIFAR10-Image-Classification","external":""},"html":"Custom CNN architecture with 1.15M parameters achieving 64.62% test accuracy. Implemented batch normalization, dropout regularization, and optimized training pipeline with data augmentation techniques.
"}},{"node":{"frontmatter":{"date":"2025-09-01","title":"Character-Level Text Generation with LSTM","tech":["PyTorch","LSTM","NLP","Python"],"github":"https://github.com/sej07/CharLevel-Text-Generation-Using-LSTM","external":""},"html":"Stacked LSTM model with 512 hidden units per layer for Shakespeare-style text generation. Trained on 1.1M character sequences with temperature sampling for controllable generation.
"}},{"node":{"frontmatter":{"date":"2024-12-01","title":"Plant Disease Classification using CNN","tech":["TensorFlow","Keras","CNNs","Computer Vision","Python"],"github":"https://github.com/sej07/Plant-Disease-Classification","external":""},"html":"Built a CNN model to classify plant leaf images into disease categories using the PlantVillage dataset (54,000+ images). Achieved 97% training accuracy and 87% validation accuracy with a custom architecture featuring data augmentation and transfer learning techniques. Developed a predictive system for real-time disease detection to support agricultural productivity.
"}},{"node":{"frontmatter":{"date":"2024-11-01","title":"Number to Words Translation using Seq2Seq","tech":["TensorFlow","Keras","LSTM","Seq2Seq","NLP"],"github":"https://github.com/sej07/Number-to-Words-Seq2Seq","external":""},"html":"Implemented a sequence-to-sequence model with stacked LSTM encoder-decoder architecture to translate numbers into their word equivalents. Built custom synthetic dataset with text preprocessing and tokenization pipeline. Trained with teacher forcing for 150 epochs, demonstrating how encoder-decoder models learn structured sequence translations without attention mechanisms.
"}},{"node":{"frontmatter":{"date":"2024-08-01","title":"News Category Classification using LSTM","tech":["TensorFlow","Keras","LSTM","NLP","Python"],"github":"https://github.com/sej07/News-Category-Classification-LSTM","external":""},"html":"Built an LSTM-based neural network to classify 200,000+ news articles into multiple categories. Implemented stacked LSTM architecture with dropout regularization and embedding layers for text representation. Achieved 56% accuracy on multi-class classification, with detailed performance analysis showing domain-specific vocabulary significantly impacts classification accuracy across categories like Politics vs. Entertainment.
"}}]}},"pageContext":{}}, "staticQueryHashes": ["1994492073","2009693873","2031412112","3825832676"]}