I built a visual ML trainer that exports to standalone PyTorch code
Visual PyTorch builder that actually exports clean code unlike Orange or KNIME.
A visual-based graph node editor for training computer vision models.
Auto-fills in_features when connecting layers, exports clean PyTorch code.
ML beginners, educators, rapid prototypers, computer vision learners
KNIME · RapidMiner · Google Teachable Machine
Visual PyTorch builder that actually exports clean code unlike Orange or KNIME.
Another visual ML builder when KNIME, Ludwig, and Azure ML Designer already exist.
Infers layer shapes from connections and exports standard PyTorch scripts.
TPU training wrapper built on torchprime; solves a real problem but torchprime already exists.
XAI-driven model improvement loop, but Weights & Biases already tracks experiments better.
Node-based image filtering running entirely locally in the browser.