Reduce foundation-model embeddings to PCs and scale to quantum range — qf_embed_reduce • edaphos

3.10.0

Reference
- Capstone — six pillars integrated
- Cross-pillar campaign
- Pilar 1 — Causal AI
- Causal backdoor on br_cerrado
- Causal on 1 095 real WoSIS profiles
- Causal-discovery trio (expert × LLM × data-driven)
- LLM-KG benchmark (Gemma 4 × GPT × Claude)
- Annotation workflow (pre-annotation + Shiny review)
- Pilar 2 — Physics-Informed ML
- Parametric ODE + Neural ODE
- Pilar 3 — 4D Pedometry
- ConvLSTM on synthetic cube
- Real MODIS + NASA POWER + EnKF
- Pilar 4 — Foundation Models
- SimCLR / MoCo embeddings
- Pilar 5 — Active Learning
- Hybrid policy + physics gate
- SoilGrids-BR live
- Pilar 6 — Quantum ML
- ZZFeatureMap + Q-KRR
- Cross-pillar bridges
- Pilar 1 × Pilar 4 — Foundation IVs
- Pilar 4 × Pilar 6 — Quantum-foundation kernel
- Unified infrastructure
- Unified uncertainty API
- Case — end-to-end Cerrado benchmark
- Releases
- v2.0.0 — Quantum kernel × Foundation (Pilar 4 × Pilar 6)
- v1.9.x — Foundation embeddings as causal IVs + Sargan/Cinelli-Hazlett
- v1.8.x — LLM benchmark infrastructure + Shiny annotation tool
- v1.7.x — Capstone vignette + native-query calibration + causal discovery
- Changelog

Takes a matrix of per-observation foundation embeddings and returns the top-n_pcs principal components rescaled into [-pi, pi], the natural input range of the ZZFeatureMap. Zero-variance columns are dropped before PCA. The rotation and scaling are stored on the return object so new observations can be projected identically.

Usage

qf_embed_reduce(embeddings, n_pcs = 8L)

Arguments

embeddings: Numeric matrix (obs x embedding-dim).
n_pcs: Integer; number of PCs to retain. Default 8L (the largest n for which quantum_kernel() stays classically simulable on a laptop).

Value

A list with components X_q (the PCs in [-pi, pi], ready for quantum_krr_fit()), rotation, variance_explained, pca_center, pca_scale, and range_min/range_max used for the pi-rescale.