TF-Recommenders & Kubernetes for flexible RecSys Model Development & Deployment

Introducing TF-Recommenders Recently, Google open sourced a Keras API for building recommender systems called TF-Recommenders. TF-Recommenders is flexible, making it easy to integrate heterogeneous signals like implicit ratings from user interactions, content embeddings, or real-time context info. This module also introduces losses specialized for ranking and retrieval which can be combined to benefit from multi-task learning. The developers emphasize the ease-of-use in research, as well as the robustness for deployment in web-scale applications....

 · 5 min · Terry Rodriguez & Salma Mayorquin

TF-Ranking and BERT for Movie Recommendations

Check out our repo for all the code referenced in this blog! Recommender systems are used by many groups to maximize the presentation of products to users. There is a variety of implementations for building recommender systems, but at their core, these systems are designed to sort a universe of items by their relevance to a user based on user information, item information, or both. One well known algorithm for solving the sorting problem is the Learn-to-Rank model, where the objective is to rank a list of examples by each item’s relevance to a particular user....

 · 6 min · Terry Rodriguez & Salma Mayorquin

IVA Pipelines with NVIDIA TLT and Deepstream SDK 5.0

We have seen applications in industries like retail, telemedicine, and robotics enabled by video analytics with machine learning. ML practitioners often leverage transfer learning with pretrained models to expedite development. Computer vision applications can benefit from using video analytics frameworks to facilitate faster iteration and experimentation. NVIDIA’s TLT toolkit and the Deepstream SDK 5.0 have made it easy to experiment with various network architectures and quickly deploy them on a NVIDIA powered device for optimized inference....

 · 3 min · Terry Rodriguez & Salma Mayorquin

Protecting Privacy With Computer Vision

Check out and contribute to our collection of data privacy resources! AI researchers developed models to identify image pixels featuring people. We apply this to promote privacy by helping you redact personally identifiable info in images. This demo is powered by Tensorflow.js! Drop an image and retrieve the redacted output without ever sending data over the internet. Click on your redacted image when it’s done to save. Consider another use case of delivery robots roaming the streets....

 · 1 min · Terry Rodriguez & Salma Mayorquin

Everybody Dance Faster

Check out the repo and the video! “Everybody Dance Now” offers a sensational demonstration in combining image-to-image translation with pose estimation to produce photo-realistic ‘do-as-i-do’ motion transfer. Researchers used roughly 20 mins of video shot at 120 fps of a subject moving through a normal range of body motion. It is also important for source and target videos to be taken from similar perspectives. Generally, this is a fixed camera angle at a third person perspective with the subject’s body occupying most of the image....

 · 7 min · Terry Rodriguez & Salma Mayorquin