Hugging Face models in Amazon Bedrock
•
10
Simon Pagezy
pagezyhf
AI & ML interests
Healthcare ML
Recent Activity
replied to
singhsidhukuldeep's
post
1 day ago
Excited to share insights from Walmart's groundbreaking semantic search system that revolutionizes e-commerce product discovery!
The team at Walmart Global Technology(the team that I am a part of 😬) has developed a hybrid retrieval system that combines traditional inverted index search with neural embedding-based search to tackle the challenging problem of tail queries in e-commerce.
Key Technical Highlights:
• The system uses a two-tower BERT architecture where one tower processes queries and another processes product information, generating dense vector representations for semantic matching.
• Product information is enriched by combining titles with key attributes like category, brand, color, and gender using special prefix tokens to help the model distinguish different attribute types.
• The neural model leverages DistilBERT with 6 layers and projects the 768-dimensional embeddings down to 256 dimensions using a linear layer, achieving optimal performance while reducing storage and computation costs.
• To improve model training, they implemented innovative negative sampling techniques combining product category matching and token overlap filtering to identify challenging negative examples.
Production Implementation Details:
• The system uses a managed ANN (Approximate Nearest Neighbor) service to enable fast retrieval, achieving 99% recall@20 with just 13ms latency.
• Query embeddings are cached with preset TTL (Time-To-Live) to reduce latency and costs in production.
• The model is exported to ONNX format and served in Java, with custom optimizations like fixed input shapes and GPU acceleration using NVIDIA T4 processors.
Results:
The system showed significant improvements in both offline metrics and live experiments, with:
- +2.84% improvement in NDCG@10 for human evaluation
- +0.54% lift in Add-to-Cart rates in live A/B testing
This is a fantastic example of how modern NLP techniques can be successfully deployed at scale to solve real-world e-
reacted
to
singhsidhukuldeep's
post
with 🤯
1 day ago
Excited to share insights from Walmart's groundbreaking semantic search system that revolutionizes e-commerce product discovery!
The team at Walmart Global Technology(the team that I am a part of 😬) has developed a hybrid retrieval system that combines traditional inverted index search with neural embedding-based search to tackle the challenging problem of tail queries in e-commerce.
Key Technical Highlights:
• The system uses a two-tower BERT architecture where one tower processes queries and another processes product information, generating dense vector representations for semantic matching.
• Product information is enriched by combining titles with key attributes like category, brand, color, and gender using special prefix tokens to help the model distinguish different attribute types.
• The neural model leverages DistilBERT with 6 layers and projects the 768-dimensional embeddings down to 256 dimensions using a linear layer, achieving optimal performance while reducing storage and computation costs.
• To improve model training, they implemented innovative negative sampling techniques combining product category matching and token overlap filtering to identify challenging negative examples.
Production Implementation Details:
• The system uses a managed ANN (Approximate Nearest Neighbor) service to enable fast retrieval, achieving 99% recall@20 with just 13ms latency.
• Query embeddings are cached with preset TTL (Time-To-Live) to reduce latency and costs in production.
• The model is exported to ONNX format and served in Java, with custom optimizations like fixed input shapes and GPU acceleration using NVIDIA T4 processors.
Results:
The system showed significant improvements in both offline metrics and live experiments, with:
- +2.84% improvement in NDCG@10 for human evaluation
- +0.54% lift in Add-to-Cart rates in live A/B testing
This is a fantastic example of how modern NLP techniques can be successfully deployed at scale to solve real-world e-
liked
a model
1 day ago
Qwen/QVQ-72B-Preview
Articles
Organizations
pagezyhf's activity
upvoted
a
collection
18 days ago
upvoted
a
collection
28 days ago
upvoted
a
collection
about 1 month ago
upvoted
a
collection
about 2 months ago
upvoted
a
collection
2 months ago
upvoted
an
article
2 months ago
Article
Inference Endpoints Changelog 🚀
By
•
•
18
upvoted
5
articles
4 months ago
Article
Hugging Face and Google partner for open AI collaboration
•
4
Article
Vision Language Models Explained
•
238
Article
Optimum-NVIDIA - Unlock blazingly fast LLM inference in just 1 line of code
•
4
Article
2024 Security Feature Highlights
•
16
Article
The 5 Most Under-Rated Tools on Hugging Face
•
86
upvoted
6
articles
5 months ago
Article
Deploy Meta Llama 3.1 405B on Google Cloud Vertex AI
•
18
Article
Serverless Inference with Hugging Face and NVIDIA NIMs
•
27
Article
Tool Use, Unified
•
70
Article
Build AI on premise with Dell Enterprise Hub
•
18
Article
XetHub is joining Hugging Face!
•
81
Article
🔥 Argilla 2.0: the data-centric tool for AI makers 🤗
By
•
•
37
upvoted
a
collection
5 months ago
upvoted
an
article
5 months ago
Article
Easily Train Models with H100 GPUs on NVIDIA DGX Cloud
•
6
upvoted
a
collection
6 months ago