PyTorch Implementations For A Series Of Deep Learning-Based Recommendation Models
-
Updated
Nov 21, 2022 - Python
PyTorch Implementations For A Series Of Deep Learning-Based Recommendation Models
Scalable Hybrid Recommender (TensorFlow + TF-IDF) - Trained on 5M/70M interactions to MAE 0.1863 / MSE 0.0727 (epoch 16, ~20 min) and shipped Flask on GKE (Docker, CI/CD GitHub Actions, HPA), −30% p95 latency, +17% relevance vs. KNN, DVC + GCS artifacts, CometML tracking.
Add a description, image, and links to the cometml topic page so that developers can more easily learn about it.
To associate your repository with the cometml topic, visit your repo's landing page and select "manage topics."