fastplotlib-scientific discovery through data visualization
summary
Claim: a GPU‑backed plotting library called fastplotlib delivers very fast, nonblocking scientific plots by building on pygfx via WGPU and modern APIs like Vulkan. Demos show live neural‑probe streaming at about 5 ms latency for 30 kHz acquisition, an interactive denoising app that updates model parameters in real time, and a 2D latent exploration flow for audio and behavior. Code runs on standard integrated GPUs, ships examples for Jupyter/Qt/GLFW, and plans include a WGSL JPEG encoder and tighter GPU ML integration.
AI-written summary · about
chapters (8)
- 0:00fastplotlib introduction
- 1:26Building new library
- 4:57Simplifying visualization
- 9:37Neuroscience use cases
- 13:53Interactive latent space
- 17:52Real-time multi-modal sync
- 22:04Key takeaways
- 25:55Q&A
related talks
21mGetting more out of Matplotlib with GR
01h 36mRAPIDS: GPU-Accelerated Data Science for PyData Users
27mBringing NVIDIA math libraries to Python scientific community
36mEnd to End Data Science Without Leaving The GPU
01h 04mVisualising Data through Pandas
38mPython for Visualization of HPC Applications
27mSpeed up Python with hardware accelerators
29mInteractive visualization for the curious