We turn your ideas into math and software
We turn your ideas into math and software
Tomi Peltola

Tomi Peltola

Tomi is a Co-founder and Machine Learning Specialist at Digisalix.Previously, Tomi has worked as Senior AI Scientist at Silo.AI, as Research Scientist at Curious AI, and as Post-doctoral Researcher with Professor Samuel Kaski in the Probabilistic Machine Learning research group at Helsinki Institute for Information Technology HIIT, Department of Computer Science, Aalto University, Finland.

Tomi Peltola's posts

Structure is Key to using Small Data sets for Intelligent Document Processing

Structure is Key to using Small Data sets for Intelligent Document Processing

In this post Digisalix machine-learning guru, Tomi, explains why appreciating different levels of structure can help reduce the data hungriness of the machine learning pipeline in intelligent document automation, leaving more time for people…
Faster bilinear interpolation gradients for PyTorch

Faster bilinear interpolation gradients for PyTorch

Shared tweaks to open-source libraries keep the whole industry moving forwards. Lessons: memory allocations can get expensive, profiler (PySpy) is your friend, and tweaking PyTorch for your specific use case can be easy.