Try DeepSparse Now
Pick an NLP or Computer Vision use case below and benchmark an optimized model to experience best-in-class CPU performance. Take one step further and train the optimized model with your private data and deploy it to your CPU infrastructure.
With the examples below, install DeepSparse, then benchmark different models like a sparse-quantized version of Hugging Face BERT-base, YOLOv5 or ResNet-50. Try SparseML, our open-source library, to transfer learn our sparse-quantized model to your dataset with a few lines of code.
See SparseZoo for other sparse models and recipes you can benchmark and prototype from.
Natural Language Processing (NLP)
| Sparse Use Case | Apply Your Data |
| Question Answering | Try It Now |
| Text Classification: Sentiment Analysis | Try It Now |
| Text Classification: Multi-Class | Try It Now |
| Text Classification: Binary | Try It Now |
| Token Classification: Named Entity Recognition | Try It Now |
Computer Vision
*Simplified end-to-end computer vision tutorials that include installable CLIs and APIs are on our short-term roadmap. For now, use the linked GitHub tutorials linked below.
| Sparse Use Case | Apply Your Data |
| Object Detection | Benchmark and Deploy Apply your Data |
| Image Classification | Benchmark and Deploy Apply your Data |