Fix single-row prediction error and enable high-speed HF transfers

Optimize HF login and silence redundant log notes

Fix SAP RPT-1 single-row prediction and add hf_xet

Move Dockerfile to root for Hugging Face detection

Fix matrix path in app.py

Fix deployment entry point and merge requirements

application file

formatted

Revert back

fix: revert stacking models to stable sklearn-native ones to avoid cloning error

feat: ensure ensembles are always visible and include TabPFN/SAP in stacking

fix: restore local execution compatibility and fix setup.py package mapping

perf: subsample to 1000 rows on HF and fix encoder consistency for correct predictions

fix: final prediction display and label JSON compatibility

fix: robust live prediction and playground UI cleanup

fix: ensure sentence-transformers is installed before pre-download

fix: pre-download SAP model weights and unify HF login

fix: handle inhomogeneous fold counts in statistical analysis

feat: finalized stable architecture with consolidated structure and single-worker mode

perf: reduce CV folds to 3 for faster benchmarking

perf: cache SAP model and fix main.py imports/sys.path

chore: add .gitignore to exclude sensitive files

fix: revert to tabpfn v2, add TABPFN_TOKEN support, and update v2 API usage

fix: downgrade tabpfn to 0.1.11, class-level weight cache, fix ravel on list

fix: pre-accept TabPFN license in Dockerfile ENV + cache file + all env var names

fix: pad single-row input for SAP RPT-1 OSS predict (np.concatenate crash)

fix: handle webapp.benchmark module path in Docker container

fix: add missing scipy dep, TabPFN weight caching, better error logging

Initial deployment of ModelMatrix-HF

initial commit