Pocket Text Lab
This is a text classification pipeline that runs entirely in the browser using WebGPU/WASM, with the Python logic executed via Pyodide. It downloads an embedding model, trains a linear SVM classifier whose hyperparameters are determined by cross-validation over a grid, and calibrates the classifier to produce reliable confidence scores. Users can paste their own data (no uploads supported) to train a model on it. Work in progress, with a blog post planned.
Shared over time
Each entry reflects what was shared on that date. Details may have changed since then.
Pocket Text Lab
This is a text classification pipeline that runs entirely in the browser using WebGPU/WASM, with the Python logic executed via Pyodide. It downloads an embedding model, trains a linear SVM classifier whose hyperparameters are determined by cross-validation over a grid, and calibrates the classifier to produce reliable confidence scores. Users can paste their own data (no uploads supported) to train a model on it. Work in progress, with a blog post planned.
Posted by @abhgh · Original comment ↗
“Given the popularity of decision models I tried to create a standard text classification pipeline that runs entirely in the browser using webgpu/wasm. Downloads an embedding model, trains the classifier (a Linear SVM for now, whose hyperparams are determined by cross validation over a grid), calibrates the classifier so that reliable confidence scores may be produced. This is in Python that is run using Pyodide.”