Graph Transforms to Quantize and Retrain Deep Neural Nets in TensorFlow
-
Updated
Dec 9, 2019 - Python
Graph Transforms to Quantize and Retrain Deep Neural Nets in TensorFlow
A image classifier that classifies whether a galaxy is spiral, elliptical, or irregular. Built using Tensorflow v2 and Google's inception v3 model.
End-to-end adaptive ML system with real-time prediction, monitoring, drift detection, and automated retraining using FastAPI and Streamlit
A self-supervised learning framework based on DINOv3 (Vision Transformer) for custom dataset retraining. This project implements training, evaluation, and visualization tools with flexible configuration options.
Retraining of LLM Models using LoRA and QLoRA
End-to-end, production-aware fraud detection system with real-time inference, delayed labels, explainability, drift monitoring, retraining, and full audit governance.
Vision Transformer Level 2 MLOps Deployment Scalable ONNX-based ViT pipeline with FastAPI, MLflow, and event-based retraining.
Scheduled ML retraining pipeline — DVC data versioning, MLflow experiment tracking, champion/challenger gating, and full audit trail. Auto-retrains weekly via GitHub Actions. Project 2/10 of MLOps portfolio.
End-to-End ML Deployment Platform with Drift Detection
Local-first MLOps system for churn prediction, model monitoring, drift detection, and controlled retraining.
My assignments from the Programming 1 module of my retraining program.
To associate your repository with the retraining topic, visit your repo's landing page and select "manage topics."