NeuralFlix

State-of-the-Art NLP Compression Research in Action: Understanding Crypto Sentiment

Presenter:

Using SOTA model compression techniques, we've been able to decrease the size of BERT NLP model by 10x, making it much more efficient to inference on commodity CPUs. See what the sparse performance looks like compared to base performance in this sentiment analysis demo where we analyzed thousands of Tweets to discover recent sentiment around cryptocurrency.

More Neural Magic Software in Action Videos

YOLOv5 on CPUs: Sparsifying to Achieve GPU-Level Performance and Tiny Footprint
YOLOv3 on the Edge: DeepSparse Engine vs. PyTorch
State-of-the-Art NLP Compression Research in Action: Understanding Crypto Sentiment
3.5x Faster NLP BERT Using a Sparsity-Aware Inference Engine on AMD Milan-X

Get more info about

State-of-the-Art NLP Compression Research in Action: Understanding Crypto Sentiment